Categories
Bibliography Data Science - Big Data Python

B07QPC8RZX

See: Data Science from Scratch: First Principles with Python 2nd Edition

Fair Use Source:

Categories
Artificial Intelligence Bibliography Data Science - Big Data Python

B07X4WW5B9

See: Machine Learning Pocket Reference: Working with Structured Data in Python 1st Edition

Fair Use Source:

Categories
Artificial Intelligence AWS Bibliography Data Science - Big Data

B0921MXC9S

See: Data Science on AWS: Implementing End-to-End, Continuous AI and Machine Learning Pipelines 1st Edition

Fair Use Source:

Categories
Azure Bibliography

Microsoft Exam Ref DP-900 Microsoft Azure Data Fundamentals

See: B08PNQDT51

Exam Ref DP-900 Microsoft Azure Data Fundamentals offers professional-level preparation that helps candidates maximize their exam performance and sharpen their skills on the job. It focuses on the specific areas of expertise modern IT professionals need to demonstrate real-world foundational knowledge of core data concepts and how they are implemented using Microsoft Azure data services. Coverage includes:

  • Describing core data concepts
  • Describing how to work with relational data on Azure
  • Describing how to work with non-relational data on Azure
  • Describing analytics workloads on Azure

See also: Microsoft Certification Exams

Fair Use Source:

Categories
Bibliography

B08QZ5KPZN

See: Applied Data Science Using PySpark: Learn the End-to-End Predictive Model-Building Cycle 1st ed. Edition

Fair Use Source:

Categories
Artificial Intelligence Cloud Data Science - Big Data History

Limited hangout

“A limited hangout or partial hangout is, according to former special assistant to the Deputy Director of the Central Intelligence Agency Victor Marchetti, “spy jargon for a favorite and frequently used gimmick of the clandestine professionals. When their veil of secrecy is shredded and they can no longer rely on a phony cover story to misinform the public, they resort to admitting—sometimes even volunteering—some of the truth while still managing to withhold the key and damaging facts in the case. The public, however, is usually so intrigued by the new information that it never thinks to pursue the matter further.”[1][2] (WP)

Modified limited hangout

“In a March 22, 1973, meeting between president Richard NixonJohn DeanJohn EhrlichmanJohn Mitchell, and H. R. Haldeman, Ehrlichman incorporated the term into a new and related one, “modified limited hangout“.[3][4]

The phrase was coined in the following exchange:[5]” (WP)

PRESIDENT: You think, you think we want to, want to go this route now? And the — let it hang out, so to speak?

DEAN: Well, it’s, it isn’t really that —
HALDEMAN: It’s a limited hang out.
DEAN: It’s a limited hang out.
EHRLICHMAN: It’s a modified limited hang out.

PRESIDENT: Well, it’s only the questions of the thing hanging out publicly or privately.

“Before this exchange, the discussion captures Nixon outlining to Dean the content of a report that Dean would create, laying out a misleading view of the role of the White House staff in events surrounding the Watergate burglary. In Ehrlichman’s words: “And the report says, ‘Nobody was involved,'”. The document would then be shared with the United States Senate Watergate Committee investigating the affair. The report would serve the administration’s goals by protecting the President, providing documentary support for his false statements should information come to light that contradicted his stated position. Further, the group discusses having information on the report leaked by those on the Committee sympathetic to the President, to put exculpatory information into the public sphere.[5]” (WP)

“The phrase has been cited as a summation of the strategy of mixing partial admissions with misinformation and resistance to further investigation, and is used in political commentary to accuse people or groups of following a Nixon-like strategy.[6]” (WP) However, this “strategy” has been used since time immemorial.

“Writing in The Washington PostMary McGrory described a statement by Pope John Paul II regarding sexual abuse by priests as a “modified, limited hangout”.[7] (WP)

See also

References

  1. ^ Victor Marchetti (August 14, 1978) The Spotlight
  2. ^ “720 F2d 631 Hunt v. Liberty Lobby Dc”. OpenJurist. 1983-11-28. Retrieved 2016-07-13.
  3. ^ Frost/Nixon: The Complete Interviews. David Frost, Richard Nixon. Paradine Television, 1977.
  4. ^ Safire, William (26 March 1989). “On Language; In Nine Little Words”New York Times. Retrieved 23 June 2013.
  5. a b “Transcript of a recording of a meeting among the president, John Dean, John Erlichman, H. R. Haldeman, and John Mitchell on March 22, 1973 from 1:57 to 3:43 p.m.” History and Politics Out Loud. Retrieved 2006-08-27.
  6. ^ Carrol, Jon (2002-05-01). “The Richard Nixon playbook”San Francisco Chronicle. Retrieved 2006-08-27.
  7. ^ McGrory, Mary (2002-04-25). “From Rome, A ‘Limited Hangout'”The Washington Post. Washington, D.C. p. A29. Retrieved 2010-04-30.

Categories

” (WP)

Sources:

Fair Use Sources:

Categories
Artificial Intelligence Cloud Data Science - Big Data Networking Software Engineering

AIoT – Artificial intelligence of Things

” (WP)

The Artificial Intelligence of Things (AIoT) is the combination of Artificial intelligence (AI) technologies with the Internet of things (IoT) infrastructure to achieve more efficient IoT operations, improve human-machine interactions and enhance data management and analytics[1] [2] [3]

See also

References

  1. ^ Ghosh, Iman (12 August 2020). “AIoT: When Artificial Intelligence Meets the Internet of Things”Visual Capitalist. Retrieved 22 September 2020.
  2. ^ Lin, Yu-Jin; Chuang, Chen-Wei; Yen, Chun-Yueh; Huang, Sheng-Hsin; Huang, Peng-Wei; Chen, Ju-Yi; Lee, Shuenn-Yuh (March 2019). “Artificial Intelligence of Things Wearable System for Cardiac Disease Detection”2019 IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS): 67–70. doi:10.1109/AICAS.2019.8771630. Retrieved 22 September 2020.
  3. ^ Chu, William Cheng-Chung; Shih, Chihhsiong; Chou, Wen-Yi; Ahamed, Sheikh Iqbal; Hsiung, Pao-Ann (November 2019). “Artificial Intelligence of Things in Sports Science: Weight Training as an Example”Computer52 (11): 52–61. doi:10.1109/MC.2019.2933772ISSN 1558-0814. Retrieved 22 September 2020.

Categories

” (WP)

Sources:

Fair Use Sources:

Categories
Data Science - Big Data Software Engineering

Ubiquitous computing – Ubicomp

” (WP)

Ubiquitous computing (or “ubicomp“) is a concept in software engineering and computer science where computing is made to appear anytime and everywhere. In contrast to desktop computingubiquitous computing can occur using any device, in any location, and in any format. A user interacts with the computer, which can exist in many different forms, including laptop computerstablets and terminals in everyday objects such as a refrigerator or a pair of glasses. The underlying technologies to support ubiquitous computing include Internet, advanced middlewareoperating systemmobile codesensorsmicroprocessors, new I/O and user interfacescomputer networks, mobile protocols, location and positioning, and new materials.

This paradigm is also described as pervasive computing,[1] ambient intelligence,[2] or “everyware”.[3] Each term emphasizes slightly different aspects. When primarily concerning the objects involved, it is also known as physical computing, the Internet of Things (IoT)haptic computing,[4] and “things that think”. Rather than propose a single definition for ubiquitous computing and for these related terms, a taxonomy of properties for ubiquitous computing has been proposed, from which different kinds or flavors of ubiquitous systems and applications can be described.[5]

Ubiquitous computing touches on distributed computingmobile computing, location computing, mobile networking, sensor networkshuman–computer interaction, context-aware smart home technologies, and artificial intelligence.

Core concepts

Ubiquitous computing is the concept of using small internet connected and inexpensive computers to help with everyday functions in an automated fashion. For example, a domestic ubiquitous computing environment might interconnect lighting and environmental controls with personal biometric monitors woven into clothing so that illumination and heating conditions in a room might be modulated, continuously and imperceptibly. Another common scenario posits refrigerators “aware” of their suitably tagged contents, able to both plan a variety of menus from the food actually on hand, and warn users of stale or spoiled food.[6]

Ubiquitous computing presents challenges across computer science: in systems design and engineering, in systems modelling, and in user interface design. Contemporary human-computer interaction models, whether command-line, menu-driven, or GUI-based, are inappropriate and inadequate to the ubiquitous case. This suggests that the “natural” interaction paradigm appropriate to a fully robust ubiquitous computing has yet to emerge – although there is also recognition in the field that in many ways we are already living in a ubicomp world (see also the main article on natural user interfaces). Contemporary devices that lend some support to this latter idea include mobile phonesdigital audio playersradio-frequency identification tags, GPS, and interactive whiteboards.

Mark Weiser proposed three basic forms for ubiquitous computing devices:[7]

  • Tabs: a wearable device that is approximately a centimeter in size
  • Pads: a hand-held device that is approximately a decimeter in size
  • Boards: an interactive larger display device that is approximately a meter in size

Ubiquitous computing devices proposed by Mark Weiser are all based around flat devices of different sizes with a visual display.[8] Expanding beyond those concepts there is a large array of other ubiquitous computing devices that could exist. Some of the additional forms that have been conceptualized are:[5]

  • Dust: miniaturized devices can be without visual output displays, e.g. micro electro-mechanical systems (MEMS), ranging from nanometres through micrometers to millimetres. See also Smart dust.
  • Skin: fabrics based upon light emitting and conductive polymers, organic computer devices, can be formed into more flexible non-planar display surfaces and products such as clothes and curtains, see OLED display. MEMS device can also be painted onto various surfaces so that a variety of physical world structures can act as networked surfaces of MEMS.
  • Clay: ensembles of MEMS can be formed into arbitrary three dimensional shapes as artefacts resembling many different kinds of physical object (see also tangible interface).

In Manuel Castells‘ book The Rise of the Network Society, Castells puts forth the concept that there is going to be a continuous evolution of computing devices. He states we will progress from stand-alone microcomputers and decentralized mainframes towards pervasive computing. Castells’ model of a pervasive computing system, uses the example of the Internet as the start of a pervasive computing system. The logical progression from that paradigm is a system where that networking logic becomes applicable in every realm of daily activity, in every location and every context. Castells envisages a system where billions of miniature, ubiquitous inter-communication devices will be spread worldwide, “like pigment in the wall paint”.

Ubiquitous computing may be seen to consist of many layers, each with their own roles, which together form a single system:

  • Layer 1: Task management layer
    • Monitors user task, context and index
    • Map user’s task to need for the services in the environment
    • To manage complex dependencies
  • Layer 2: Environment management layer
    • To monitor a resource and its capabilities
    • To map service need, user level states of specific capabilities
  • Layer 3: Environment layer
    • To monitor a relevant resource
    • To manage reliability of the resources

History

Mark Weiser coined the phrase “ubiquitous computing” around 1988, during his tenure as Chief Technologist of the Xerox Palo Alto Research Center (PARC). Both alone and with PARC Director and Chief Scientist John Seely Brown, Weiser wrote some of the earliest papers on the subject, largely defining it and sketching out its major concerns.[7][9][10]

Recognizing the effects of extending processing power

Recognizing that the extension of processing power into everyday scenarios would necessitate understandings of social, cultural and psychological phenomena beyond its proper ambit, Weiser was influenced by many fields outside computer science, including “philosophyphenomenologyanthropologypsychologyand sociology of science “. He was explicit about “the humanistic origins of the invisible ideal'”,[10] referencing as well the ironically dystopian Philip K. Dick novel Ubik.

Andy Hopper from Cambridge University UK proposed and demonstrated the concept of “Teleporting” – where applications follow the user wherever he/she moves.

Roy Want, while a researcher and student working under Andy Hopper at Cambridge University, worked on the “Active Badge System”, which is an advanced location computing system where personal mobility that is merged with computing.

Bill Schilit (now at Google) also did some earlier work in this topic, and participated in the early Mobile Computing workshop held in Santa Cruz in 1996.

Ken Sakamura of the University of TokyoJapan leads the Ubiquitous Networking Laboratory (UNL), Tokyo as well as the T-Engine Forum. The joint goal of Sakamura’s Ubiquitous Networking specification and the T-Engine forum, is to enable any everyday device to broadcast and receive information.[11][12]

MIT has also contributed significant research in this field, notably Things That Think consortium (directed by Hiroshi IshiiJoseph A. Paradiso and Rosalind Picard) at the Media Lab[13] and the CSAIL effort known as Project Oxygen.[14] Other major contributors include University of Washington‘s Ubicomp Lab (directed by Shwetak Patel), Dartmouth College‘s DartNets LabGeorgia Tech‘s College of ComputingCornell University‘s People Aware Computing LabNYU‘s Interactive Telecommunications ProgramUC Irvine‘s Department of Informatics, Microsoft ResearchIntel Research and Equator,[15] Ajou University UCRi & CUS.[16]

Examples

One of the earliest ubiquitous systems was artist Natalie Jeremijenko‘s “Live Wire”, also known as “Dangling String”, installed at Xerox PARC during Mark Weiser’s time there.[17] This was a piece of string attached to a stepper motor and controlled by a LAN connection; network activity caused the string to twitch, yielding a peripherally noticeable indication of traffic. Weiser called this an example of calm technology.[18]

A present manifestation of this trend is the widespread diffusion of mobile phones. Many mobile phones support high speed data transmission, video services, and other services with powerful computational ability. Although these mobile devices are not necessarily manifestations of ubiquitous computing, there are examples, such as Japan’s Yaoyorozu (“Eight Million Gods”) Project in which mobile devices, coupled with radio frequency identification tags demonstrate that ubiquitous computing is already present in some form.[19]

Ambient Devices has produced an “orb”, a “dashboard”, and a “weather beacon“: these decorative devices receive data from a wireless network and report current events, such as stock prices and the weather, like the Nabaztag produced by Violet Snowden.

The Australian futurist Mark Pesce has produced a highly configurable 52-LED LAMP enabled lamp which uses Wi-Fi named MooresCloud after Gordon Moore.[20]

The Unified Computer Intelligence Corporation launched a device called Ubi – The Ubiquitous Computer designed to allow voice interaction with the home and provide constant access to information.[21]

Ubiquitous computing research has focused on building an environment in which computers allow humans to focus attention on select aspects of the environment and operate in supervisory and policy-making roles. Ubiquitous computing emphasizes the creation of a human computer interface that can interpret and support a user’s intentions. For example, MIT’s Project Oxygen seeks to create a system in which computation is as pervasive as air:

In the future, computation will be human centered. It will be freely available everywhere, like batteries and power sockets, or oxygen in the air we breathe…We will not need to carry our own devices around with us. Instead, configurable generic devices, either handheld or embedded in the environment, will bring computation to us, whenever we need it and wherever we might be. As we interact with these “anonymous” devices, they will adopt our information personalities. They will respect our desires for privacy and security. We won’t have to type, click, or learn new computer jargon. Instead, we’ll communicate naturally, using speech and gestures that describe our intent…[22]

This is a fundamental transition that does not seek to escape the physical world and “enter some metallic, gigabyte-infested cyberspace” but rather brings computers and communications to us, making them “synonymous with the useful tasks they perform”.[19]

Network robots link ubiquitous networks with robots, contributing to the creation of new lifestyles and solutions to address a variety of social problems including the aging of population and nursing care.[23]

Issues

Privacy is easily the most often-cited criticism of ubiquitous computing (ubicomp), and may be the greatest barrier to its long-term success.[24]

Public policy problems are often “preceded by long shadows, long trains of activity”, emerging slowly, over decades or even the course of a century. There is a need for a long-term view to guide policy decision making, as this will assist in identifying long-term problems or opportunities related to the ubiquitous computing environment. This information can reduce uncertainty and guide the decisions of both policy makers and those directly involved in system development (Wedemeyer et al. 2001). One important consideration is the degree to which different opinions form around a single problem. Some issues may have strong consensus about their importance, even if there are great differences in opinion regarding the cause or solution. For example, few people will differ in their assessment of a highly tangible problem with physical impact such as terrorists using new weapons of mass destruction to destroy human life. The problem statements outlined above that address the future evolution of the human species or challenges to identity have clear cultural or religious implications and are likely to have greater variance in opinion about them.[19]

Ubiquitous computing research centres

This is a list of notable institutions who claim to have a focus on Ubiquitous computing sorted by country:Canada

Topological Media Lab, Concordia University, CanadaFinland

Community Imaging Group, University of Oulu, FinlandGermany

Tele cooperation Office (TECO), Karlsruhe Institute of Technology, GermanyIndia

Ubiquitous Computing Research Resource Centre (UCRC), Centre for Development of Advanced Computing[25]Pakistan

Centre for Research in Ubiquitous Computing (CRUC), Karachi, Pakistan.Sweden

Mobile Life Centre, Stockholm UniversityUnited Kingdom

Mixed Reality Lab, University of Nottingham

See also

References

  1. ^ Nieuwdorp, E. (2007). “The pervasive discourse”. Computers in Entertainment5(2): 13. doi:10.1145/1279540.1279553S2CID 17759896.
  2. ^ Hansmann, Uwe (2003). Pervasive Computing: The Mobile World. Springer. ISBN 978-3-540-00218-5.
  3. ^ Greenfield, Adam (2006). Everyware: The Dawning Age of Ubiquitous Computing. New Riders. pp. 11–12. ISBN 978-0-321-38401-0.
  4. ^ “World Haptics Conferences”. Haptics Technical Committee. Archived from the original on 16 November 2011.
  5. a b Poslad, Stefan (2009). Ubiquitous Computing Smart Devices, Smart Environments and Smart Interaction (PDF). Wiley. ISBN 978-0-470-03560-3.
  6. ^ Kang, Byeong-Ho (January 2007). “Ubiquitous Computing Environment Threats and Defensive Measures”International Journal of Multimedia and Ubiquitous Engineering2 (1): 47–60. Retrieved 2019-03-22.
  7. a b Weiser, Mark (1991). “The Computer for the 21st Century”. Archived from the original on 22 October 2014.
  8. ^ Weiser, Mark (March 23, 1993). “Some Computer Science Issues in Ubiquitous Computing”. CACM. Retrieved May 28, 2019.
  9. ^ Weiser, M.; Gold, R.; Brown, J.S. (1999-05-11). “Ubiquitous computing”. Archived from the original on 10 March 2009.
  10. a b Weiser, Mark (17 March 1996). “Ubiquitous computing”. Archived from the original on 2 June 2018.
  11. ^ Krikke, J (2005). “T-Engine: Japan’s ubiquitous computing architecture is ready for prime time”. IEEE Pervasive Computing4 (2): 4–9. doi:10.1109/MPRV.2005.40S2CID 11365911.
  12. ^ “T-Engine Forum Summary”. T-engine.org. Archived from the original on 21 October 2018. Retrieved 25 August 2011.
  13. ^ “MIT Media Lab – Things That Think Consortium”MIT. Retrieved 2007-11-03.
  14. ^ “MIT Project Oxygen: Overview”MIT. Retrieved 2007-11-03.
  15. ^ “Equator”UCL. Retrieved 2009-11-19.
  16. ^ “Center of excellence for Ubiquitous System” (in Korean). CUS. Archived from the original on 2 October 2011.
  17. ^ Weiser, Mark (2017-05-03). “Designing Calm Technology”. Retrieved May 27,2019.
  18. ^ Weiser, Mark; Gold, Rich; Brown, John Seely (1999). “The Origins of Ubiquitous Computing Research at PARC in the Late 1980s”IBM Systems Journal38 (4): 693. doi:10.1147/sj.384.0693S2CID 38805890.
  19. a b c Winter, Jenifer (December 2008). “Emerging Policy Problems Related to Ubiquitous Computing: Negotiating Stakeholders’ Visions of the Future”. Knowledge, Technology & Policy21 (4): 191–203. doi:10.1007/s12130-008-9058-4hdl:10125/63534S2CID 109339320.
  20. ^ Fingas, Jon (13 October 2012). “MooresCloud Light runs Linux, puts LAMP on your lamp (video)”. Engadget.com. Retrieved 22 March 2019.
  21. ^ “Ubi Cloud”. Theubi.com. Archived from the original on 2 January 2015.
  22. ^ “MIT Project Oxygen: Overview”. Archived from the original on July 5, 2004.
  23. ^ “Network Robot Forum”. Archived from the original on October 24, 2007.
  24. ^ Hong, Jason I.; Landay, James A. (June 2004). “An architecture for privacy-sensitive ubiquitous computing” (PDF). Proceedings of the 2nd international conference on Mobile systems, applications, and services – MobiSYS ’04. pp. 177=189. doi:10.1145/990064.990087ISBN 1581137931S2CID 3776760.
  25. ^ “Ubiquitous Computing Projects”Department of Electronics & Information Technology (DeitY). Ministry of Communications & IT, Government of India. Archived from the original on 2015-07-07. Retrieved 2015-07-07.

Further reading

External links

Wikimedia Commons has media related to Ubiquitous computing.

Categories

” (WP)

Sources:

Fair Use Sources:

Categories
Artificial Intelligence Cloud Data Science - Big Data Software Engineering

AI – Artificial intelligence

“AI” redirects here. For other uses, see AI (disambiguation) and Artificial intelligence (disambiguation).

See also: Artificial Intelligence (AI) Coined – 1955 AD

” (WP)

Artificial intelligence (AI) is intelligence demonstrated by machines, unlike the natural intelligence displayed by humans and animals, which involves consciousness and emotionality. The distinction between the former and the latter categories is often revealed by the acronym chosen. ‘Strong’ AI is usually labelled as artificial general intelligence (AGI) while attempts to emulate ‘natural’ intelligence have been called artificial biological intelligence (ABI). Leading AI textbooks define the field as the study of “intelligent agents“: any device that perceives its environment and takes actions that maximize its chance of successfully achieving its goals.[3] Colloquially, the term “artificial intelligence” is often used to describe machines that mimic “cognitive” functions that humans associate with the human mind, such as “learning” and “problem solving”.[4]

As machines become increasingly capable, tasks considered to require “intelligence” are often removed from the definition of AI, a phenomenon known as the AI effect.[5] A quip in Tesler’s Theorem says “AI is whatever hasn’t been done yet.”[6] For instance, optical character recognition is frequently excluded from things considered to be AI,[7] having become a routine technology.[8] Modern machine capabilities generally classified as AI include successfully understanding human speech,[9] competing at the highest level in strategic game systems (such as chess and Go),[10] and also imperfect-information games like poker,[11] self-driving cars, intelligent routing in content delivery networks, and military simulations.[12]

Artificial intelligence was founded as an academic discipline in 1955, and in the years since has experienced several waves of optimism,[13][14] followed by disappointment and the loss of funding (known as an “AI winter“),[15][16] followed by new approaches, success and renewed funding.[14][17] After AlphaGo successfully defeated a professional Go player in 2015, artificial intelligence once again attracted widespread global attention.[18] For most of its history, AI research has been divided into sub-fields that often fail to communicate with each other.[19] These sub-fields are based on technical considerations, such as particular goals (e.g. “robotics” or “machine learning“),[20] the use of particular tools (“logic” or artificial neural networks), or deep philosophical differences.[23][24][25] Sub-fields have also been based on social factors (particular institutions or the work of particular researchers).[19]

The traditional problems (or goals) of AI research include reasoningknowledge representationplanninglearningnatural language processingperception and the ability to move and manipulate objects.[20] AGI is among the field’s long-term goals.[26] Approaches include statistical methodscomputational intelligence, and traditional symbolic AI. Many tools are used in AI, including versions of search and mathematical optimization, artificial neural networks, and methods based on statistics, probability and economics. The AI field draws upon computer scienceinformation engineeringmathematicspsychologylinguisticsphilosophy, and many other fields.

The field was founded on the assumption that human intelligence “can be so precisely described that a machine can be made to simulate it”.[27] This raises philosophical arguments about the mind and the ethics of creating artificial beings endowed with human-like intelligence. These issues have been explored by mythfiction and philosophy since antiquity.[32] Some people also consider AI to be a danger to humanity if it progresses unabated.[33][34] Others believe that AI, unlike previous technological revolutions, will create a risk of mass unemployment.[35]

In the twenty-first century, AI techniques have experienced a resurgence following concurrent advances in computer power, large amounts of data, and theoretical understanding; and AI techniques have become an essential part of the technology industry, helping to solve many challenging problems in computer science, software engineering and operations research.[36][17]

History

Main articles: History of artificial intelligence and Timeline of artificial intelligenceSilver didrachma from Crete depicting Talos, an ancient mythical automaton with artificial intelligence

Thought-capable artificial beings appeared as storytelling devices in antiquity,[37] and have been common in fiction, as in Mary Shelley‘s Frankenstein or Karel Čapek‘s R.U.R.[38] These characters and their fates raised many of the same issues now discussed in the ethics of artificial intelligence.[32]

The study of mechanical or “formal” reasoning began with philosophers and mathematicians in antiquity. The study of mathematical logic led directly to Alan Turing‘s theory of computation, which suggested that a machine, by shuffling symbols as simple as “0” and “1”, could simulate any conceivable act of mathematical deduction. This insight, that digital computers can simulate any process of formal reasoning, is known as the Church–Turing thesis.[39] Along with concurrent discoveries in neurobiologyinformation theory and cybernetics, this led researchers to consider the possibility of building an electronic brain. Turing proposed changing the question from whether a machine was intelligent, to “whether or not it is possible for machinery to show intelligent behaviour”.[40] The first work that is now generally recognized as AI was McCullouch and Pitts‘ 1943 formal design for Turing-complete “artificial neurons”.[41]

The field of AI research was born at a workshop at Dartmouth College in 1956,[42] where the term “Artificial Intelligence” was coined by John McCarthy to distinguish the field from cybernetics and escape the influence of the cyberneticist Norbert Wiener.[43] Attendees Allen Newell (CMU), Herbert Simon (CMU), John McCarthy (MIT), Marvin Minsky (MIT) and Arthur Samuel (IBM) became the founders and leaders of AI research.[44] They and their students produced programs that the press described as “astonishing”:[45] computers were learning checkers strategies (c. 1954)[46] (and by 1959 were reportedly playing better than the average human),[47] solving word problems in algebra, proving logical theorems (Logic Theorist, first run c. 1956) and speaking English.[48] By the middle of the 1960s, research in the U.S. was heavily funded by the Department of Defense[49] and laboratories had been established around the world.[50] AI’s founders were optimistic about the future: Herbert Simon predicted, “machines will be capable, within twenty years, of doing any work a man can do”. Marvin Minsky agreed, writing, “within a generation … the problem of creating ‘artificial intelligence’ will substantially be solved”.[13]

They failed to recognize the difficulty of some of the remaining tasks. Progress slowed and in 1974, in response to the criticism of Sir James Lighthill[51] and ongoing pressure from the US Congress to fund more productive projects, both the U.S. and British governments cut off exploratory research in AI. The next few years would later be called an “AI winter“,[15] a period when obtaining funding for AI projects was difficult.

In the early 1980s, AI research was revived by the commercial success of expert systems,[52] a form of AI program that simulated the knowledge and analytical skills of human experts. By 1985, the market for AI had reached over a billion dollars. At the same time, Japan’s fifth generation computer project inspired the U.S and British governments to restore funding for academic research.[14] However, beginning with the collapse of the Lisp Machine market in 1987, AI once again fell into disrepute, and a second, longer-lasting hiatus began.[16]

The development of metal–oxide–semiconductor (MOS) very-large-scale integration (VLSI), in the form of complementary MOS (CMOS) transistor technology, enabled the development of practical artificial neural network (ANN) technology in the 1980s. A landmark publication in the field was the 1989 book Analog VLSI Implementation of Neural Systems by Carver A. Mead and Mohammed Ismail.[53]

In the late 1990s and early 21st century, AI began to be used for logistics, data miningmedical diagnosis and other areas.[36] The success was due to increasing computational power (see Moore’s law and transistor count), greater emphasis on solving specific problems, new ties between AI and other fields (such as statisticseconomics and mathematics), and a commitment by researchers to mathematical methods and scientific standards.[54] Deep Blue became the first computer chess-playing system to beat a reigning world chess champion, Garry Kasparov, on 11 May 1997.[55]

In 2011, in a Jeopardy! quiz show exhibition match, IBM‘s question answering systemWatson, defeated the two greatest Jeopardy! champions, Brad Rutter and Ken Jennings, by a significant margin.[56] Faster computers, algorithmic improvements, and access to large amounts of data enabled advances in machine learning and perception; data-hungry deep learning methods started to dominate accuracy benchmarks around 2012.[57] The Kinect, which provides a 3D body–motion interface for the Xbox 360 and the Xbox One, uses algorithms that emerged from lengthy AI research[58] as do intelligent personal assistants in smartphones.[59] In March 2016, AlphaGo won 4 out of 5 games of Go in a match with Go champion Lee Sedol, becoming the first computer Go-playing system to beat a professional Go player without handicaps.[10][60] In the 2017 Future of Go SummitAlphaGo won a three-game match with Ke Jie,[61] who at the time continuously held the world No. 1 ranking for two years.[62][63] Deep Blue‘s Murray Campbell called AlphaGo’s victory “the end of an era… board games are more or less done[64] and it’s time to move on.”[65] This marked the completion of a significant milestone in the development of Artificial Intelligence as Go is a relatively complex game, more so than Chess. AlphaGo was later improved, generalized to other games like chess, with AlphaZero;[66] and MuZero[67] to play many different video games, that were previously handled separately,[68] in addition to board games. Other programs handle imperfect-information games; such as for poker at a superhuman level, Pluribus (poker bot)[69] and Cepheus (poker bot).[11] See: General game playing.

According to Bloomberg’s Jack Clark, 2015 was a landmark year for artificial intelligence, with the number of software projects that use AI within Google increased from a “sporadic usage” in 2012 to more than 2,700 projects. Clark also presents factual data indicating the improvements of AI since 2012 supported by lower error rates in image processing tasks.[70] He attributes this to an increase in affordable neural networks, due to a rise in cloud computing infrastructure and to an increase in research tools and datasets.[17] Other cited examples include Microsoft’s development of a Skype system that can automatically translate from one language to another and Facebook’s system that can describe images to blind people.[70] In a 2017 survey, one in five companies reported they had “incorporated AI in some offerings or processes”.[71][72] Around 2016, China greatly accelerated its government funding; given its large supply of data and its rapidly increasing research output, some observers believe it may be on track to becoming an “AI superpower”.[73][74]

By 2020, Natural Language Processing systems such as the enormous GPT-3 (then by far the largest artificial neural network) were matching human performance on pre-existing benchmarks, albeit without the system attaining commonsense understanding of the contents of the benchmarks.[75] DeepMind’s AlphaFold 2 (2020) demonstrated the ability to determine, in hours rather than months, the 3D structure of a protein. Facial recognition advanced to where, under some circumstances, some systems claim to have a 99% accuracy rate.[76]

Basics

Computer science defines AI research as the study of “intelligent agents“: any device that perceives its environment and takes actions that maximize its chance of successfully achieving its goals.[3] A more elaborate definition characterizes AI as “a system’s ability to correctly interpret external data, to learn from such data, and to use those learnings to achieve specific goals and tasks through flexible adaptation.”[77]

A typical AI analyzes its environment and takes actions that maximize its chance of success.[3] An AI’s intended utility function (or goal) can be simple (“1 if the AI wins a game of Go, 0 otherwise”) or complex (“Perform actions mathematically similar to ones that succeeded in the past”). Goals can be explicitly defined or induced. If the AI is programmed for “reinforcement learning“, goals can be implicitly induced by rewarding some types of behavior or punishing others.[a] Alternatively, an evolutionary system can induce goals by using a “fitness function” to mutate and preferentially replicate high-scoring AI systems, similar to how animals evolved to innately desire certain goals such as finding food.[78] Some AI systems, such as nearest-neighbor, instead of reason by analogy, these systems are not generally given goals, except to the degree that goals are implicit in their training data.[79] Such systems can still be benchmarked if the non-goal system is framed as a system whose “goal” is to successfully accomplish its narrow classification task.[80]

AI often revolves around the use of algorithms. An algorithm is a set of unambiguous instructions that a mechanical computer can execute.[b] A complex algorithm is often built on top of other, simpler, algorithms. A simple example of an algorithm is the following (optimal for first player) recipe for play at tic-tac-toe:[81]

  1. If someone has a “threat” (that is, two in a row), take the remaining square. Otherwise,
  2. if a move “forks” to create two threats at once, play that move. Otherwise,
  3. take the center square if it is free. Otherwise,
  4. if your opponent has played in a corner, take the opposite corner. Otherwise,
  5. take an empty corner if one exists. Otherwise,
  6. take any empty square.

Many AI algorithms are capable of learning from data; they can enhance themselves by learning new heuristics (strategies, or “rules of thumb”, that have worked well in the past), or can themselves write other algorithms. Some of the “learners” described below, including Bayesian networks, decision trees, and nearest-neighbor, could theoretically, (given infinite data, time, and memory) learn to approximate any function, including which combination of mathematical functions would best describe the world.[citation needed] These learners could therefore derive all possible knowledge, by considering every possible hypothesis and matching them against the data. In practice, it is seldom possible to consider every possibility, because of the phenomenon of “combinatorial explosion“, where the time needed to solve a problem grows exponentially. Much of AI research involves figuring out how to identify and avoid considering a broad range of possibilities unlikely to be beneficial.[82][83] For example, when viewing a map and looking for the shortest driving route from Denver to New York in the East, one can in most cases skip looking at any path through San Francisco or other areas far to the West; thus, an AI wielding a pathfinding algorithm like A* can avoid the combinatorial explosion that would ensue if every possible route had to be ponderously considered.[84]

The earliest (and easiest to understand) approach to AI was symbolism (such as formal logic): “If an otherwise healthy adult has a fever, then they may have influenza“. A second, more general, approach is Bayesian inference: “If the current patient has a fever, adjust the probability they have influenza in such-and-such way”. The third major approach, extremely popular in routine business AI applications, are analogizers such as SVM and nearest-neighbor: “After examining the records of known past patients whose temperature, symptoms, age, and other factors mostly match the current patient, X% of those patients turned out to have influenza”. A fourth approach is harder to intuitively understand, but is inspired by how the brain’s machinery works: the artificial neural network approach uses artificial “neurons” that can learn by comparing itself to the desired output and altering the strengths of the connections between its internal neurons to “reinforce” connections that seemed to be useful. These four main approaches can overlap with each other and with evolutionary systems; for example, neural nets can learn to make inferences, to generalize, and to make analogies. Some systems implicitly or explicitly use multiple of these approaches, alongside many other AI and non-AI algorithms; the best approach is often different depending on the problem.[85][86]

Learning algorithms work on the basis that strategies, algorithms, and inferences that worked well in the past are likely to continue working well in the future. These inferences can be obvious, such as “since the sun rose every morning for the last 10,000 days, it will probably rise tomorrow morning as well”. They can be nuanced, such as “X% of families have geographically separate species with color variants, so there is a Y% chance that undiscovered black swans exist”. Learners also work on the basis of “Occam’s razor“: The simplest theory that explains the data is the likeliest. Therefore, according to Occam’s razor principle, a learner must be designed such that it prefers simpler theories to complex theories, except in cases where the complex theory is proven substantially better.The blue line could be an example of overfitting a linear function due to random noise.

Settling on a bad, overly complex theory gerrymandered to fit all the past training data is known as overfitting. Many systems attempt to reduce overfitting by rewarding a theory in accordance with how well it fits the data, but penalizing the theory in accordance with how complex the theory is.[87] Besides classic overfitting, learners can also disappoint by “learning the wrong lesson”. A toy example is that an image classifier trained only on pictures of brown horses and black cats might conclude that all brown patches are likely to be horses.[88] A real-world example is that, unlike humans, current image classifiers often don’t primarily make judgments from the spatial relationship between components of the picture, and they learn relationships between pixels that humans are oblivious to, but that still correlate with images of certain types of real objects. Modifying these patterns on a legitimate image can result in “adversarial” images that the system misclassifies.[c][89][90]A self-driving car system may use a neural network to determine which parts of the picture seem to match previous training images of pedestrians, and then model those areas as slow-moving but somewhat unpredictable rectangular prisms that must be avoided.

Compared with humans, existing AI lacks several features of human “commonsense reasoning“; most notably, humans have powerful mechanisms for reasoning about “naïve physics” such as space, time, and physical interactions. This enables even young children to easily make inferences like “If I roll this pen off a table, it will fall on the floor”. Humans also have a powerful mechanism of “folk psychology” that helps them to interpret natural-language sentences such as “The city councilmen refused the demonstrators a permit because they advocated violence” (A generic AI has difficulty discerning whether the ones alleged to be advocating violence are the councilmen or the demonstrators[91][92][93]). This lack of “common knowledge” means that AI often makes different mistakes than humans make, in ways that can seem incomprehensible. For example, existing self-driving cars cannot reason about the location nor the intentions of pedestrians in the exact way that humans do, and instead must use non-human modes of reasoning to avoid accidents.[94][95][96]

Challenges

The cognitive capabilities of current architectures are very limited, using only a simplified version of what intelligence is really capable of. For instance, the human mind has come up with ways to reason beyond measure and logical explanations to different occurrences in life. What would have been otherwise straightforward, an equivalently difficult problem may be challenging to solve computationally as opposed to using the human mind. This gives rise to two classes of models: structuralist and functionalist. The structural models aim to loosely mimic the basic intelligence operations of the mind such as reasoning and logic. The functional model refers to the correlating data to its computed counterpart.[97]

The overall research goal of artificial intelligence is to create technology that allows computers and machines to function in an intelligent manner. The general problem of simulating (or creating) intelligence has been broken down into sub-problems. These consist of particular traits or capabilities that researchers expect an intelligent system to display. The traits described below have received the most attention.[20]

Reasoning, problem solving

Early researchers developed algorithms that imitated step-by-step reasoning that humans use when they solve puzzles or make logical deductions.[98] By the late 1980s and 1990s, AI research had developed methods for dealing with uncertain or incomplete information, employing concepts from probability and economics.[99]

These algorithms proved to be insufficient for solving large reasoning problems because they experienced a “combinatorial explosion”: they became exponentially slower as the problems grew larger.[82] Even humans rarely use the step-by-step deduction that early AI research could model. They solve most of their problems using fast, intuitive judgments.[100]

Knowledge representation

An ontology represents knowledge as a set of concepts within a domain and the relationships between those concepts.Main articles: Knowledge representation and Commonsense knowledge

Knowledge representation[101] and knowledge engineering[102] are central to classical AI research. Some “expert systems” attempt to gather explicit knowledge possessed by experts in some narrow domain. In addition, some projects attempt to gather the “commonsense knowledge” known to the average person into a database containing extensive knowledge about the world. Among the things a comprehensive commonsense knowledge base would contain are: objects, properties, categories and relations between objects;[103] situations, events, states and time;[104] causes and effects;[105] knowledge about knowledge (what we know about what other people know);[106] and many other, less well researched domains. A representation of “what exists” is an ontology: the set of objects, relations, concepts, and properties formally described so that software agents can interpret them. The semantics of these are captured as description logic concepts, roles, and individuals, and typically implemented as classes, properties, and individuals in the Web Ontology Language.[107] The most general ontologies are called upper ontologies, which attempt to provide a foundation for all other knowledge[108] by acting as mediators between domain ontologies that cover specific knowledge about a particular knowledge domain (field of interest or area of concern). Such formal knowledge representations can be used in content-based indexing and retrieval,[109] scene interpretation,[110] clinical decision support,[111] knowledge discovery (mining “interesting” and actionable inferences from large databases),[112] and other areas.[113]

Among the most difficult problems in knowledge representation are:Default reasoning and the qualification problemMany of the things people know take the form of “working assumptions”. For example, if a bird comes up in conversation, people typically picture a fist-sized animal that sings and flies. None of these things are true about all birds. John McCarthy identified this problem in 1969[114] as the qualification problem: for any commonsense rule that AI researchers care to represent, there tend to be a huge number of exceptions. Almost nothing is simply true or false in the way that abstract logic requires. AI research has explored a number of solutions to this problem.[115]Breadth of commonsense knowledgeThe number of atomic facts that the average person knows is very large. Research projects that attempt to build a complete knowledge base of commonsense knowledge (e.g., Cyc) require enormous amounts of laborious ontological engineering—they must be built, by hand, one complicated concept at a time.[116]Subsymbolic form of some commonsense knowledgeMuch of what people know is not represented as “facts” or “statements” that they could express verbally. For example, a chess master will avoid a particular chess position because it “feels too exposed”[117] or an art critic can take one look at a statue and realize that it is a fake.[118] These are non-conscious and sub-symbolic intuitions or tendencies in the human brain.[119] Knowledge like this informs, supports and provides a context for symbolic, conscious knowledge. As with the related problem of sub-symbolic reasoning, it is hoped that situated AIcomputational intelligence, or statistical AI will provide ways to represent this knowledge.[119]

Planning

hierarchical control system is a form of control system in which a set of devices and governing software is arranged in a hierarchy.Main article: Automated planning and scheduling

Intelligent agents must be able to set goals and achieve them.[120] They need a way to visualize the future—a representation of the state of the world and be able to make predictions about how their actions will change it—and be able to make choices that maximize the utility (or “value”) of available choices.[121]

In classical planning problems, the agent can assume that it is the only system acting in the world, allowing the agent to be certain of the consequences of its actions.[122] However, if the agent is not the only actor, then it requires that the agent can reason under uncertainty. This calls for an agent that can not only assess its environment and make predictions but also evaluate its predictions and adapt based on its assessment.[123]

Multi-agent planning uses the cooperation and competition of many agents to achieve a given goal. Emergent behavior such as this is used by evolutionary algorithms and swarm intelligence.[124]

Learning

Main article: Machine learningFor this project the AI had to find the typical patterns in the colors and brushstrokes of Renaissance painter Raphael. The portrait shows the face of the actress Ornella Muti, “painted” by AI in the style of Raphael.

Machine learning (ML), a fundamental concept of AI research since the field’s inception,[d] is the study of computer algorithms that improve automatically through experience.[e][127]

Unsupervised learning is the ability to find patterns in a stream of input, without requiring a human to label the inputs first. Supervised learning includes both classification and numerical regression, which requires a human to label the input data first. Classification is used to determine what category something belongs in, and occurs after a program sees a number of examples of things from several categories. Regression is the attempt to produce a function that describes the relationship between inputs and outputs and predicts how the outputs should change as the inputs change.[127] Both classifiers and regression learners can be viewed as “function approximators” trying to learn an unknown (possibly implicit) function; for example, a spam classifier can be viewed as learning a function that maps from the text of an email to one of two categories, “spam” or “not spam”. Computational learning theory can assess learners by computational complexity, by sample complexity (how much data is required), or by other notions of optimization.[128] In reinforcement learning[129] the agent is rewarded for good responses and punished for bad ones. The agent uses this sequence of rewards and punishments to form a strategy for operating in its problem space.

Natural language processing

parse tree represents the syntactic structure of a sentence according to some formal grammar.Main article: Natural language processing

Natural language processing[130] (NLP) allows machines to read and understand human language. A sufficiently powerful natural language processing system would enable natural-language user interfaces and the acquisition of knowledge directly from human-written sources, such as newswire texts. Some straightforward applications of natural language processing include information retrievaltext miningquestion answering[131] and machine translation.[132] Many current approaches use word co-occurrence frequencies to construct syntactic representations of text. “Keyword spotting” strategies for search are popular and scalable but dumb; a search query for “dog” might only match documents with the literal word “dog” and miss a document with the word “poodle”. “Lexical affinity” strategies use the occurrence of words such as “accident” to assess the sentiment of a document. Modern statistical NLP approaches can combine all these strategies as well as others, and often achieve acceptable accuracy at the page or paragraph level. Beyond semantic NLP, the ultimate goal of “narrative” NLP is to embody a full understanding of commonsense reasoning.[133] By 2019, transformer-based deep learning architectures could generate coherent text.[134]

Perception

Main articles: Machine perceptionComputer vision, and Speech recognitionFeature detection (pictured: edge detection) helps AI compose informative abstract structures out of raw data.

Machine perception[135] is the ability to use input from sensors (such as cameras (visible spectrum or infrared), microphones, wireless signals, and active lidar, sonar, radar, and tactile sensors) to deduce aspects of the world. Applications include speech recognition,[136] facial recognition, and object recognition.[137] Computer vision is the ability to analyze visual input. Such input is usually ambiguous; a giant, fifty-meter-tall pedestrian far away may produce the same pixels as a nearby normal-sized pedestrian, requiring the AI to judge the relative likelihood and reasonableness of different interpretations, for example by using its “object model” to assess that fifty-meter pedestrians do not exist.[138]

Motion and manipulation

Main article: Robotics

AI is heavily used in robotics.[139] Advanced robotic arms and other industrial robots, widely used in modern factories, can learn from experience how to move efficiently despite the presence of friction and gear slippage.[140] A modern mobile robot, when given a small, static, and visible environment, can easily determine its location and map its environment; however, dynamic environments, such as (in endoscopy) the interior of a patient’s breathing body, pose a greater challenge. Motion planning is the process of breaking down a movement task into “primitives” such as individual joint movements. Such movement often involves compliant motion, a process where movement requires maintaining physical contact with an object.[141][142][143] Moravec’s paradox generalizes that low-level sensorimotor skills that humans take for granted are, counterintuitively, difficult to program into a robot; the paradox is named after Hans Moravec, who stated in 1988 that “it is comparatively easy to make computers exhibit adult level performance on intelligence tests or playing checkers, and difficult or impossible to give them the skills of a one-year-old when it comes to perception and mobility”.[144][145] This is attributed to the fact that, unlike checkers, physical dexterity has been a direct target of natural selection for millions of years.[146]

Social intelligence

Main article: Affective computingKismet, a robot with rudimentary social skills[147]

Moravec’s paradox can be extended to many forms of social intelligence.[148][149] Distributed multi-agent coordination of autonomous vehicles remains a difficult problem.[150] Affective computing is an interdisciplinary umbrella that comprises systems which recognize, interpret, process, or simulate human affects.[151][152][153] Moderate successes related to affective computing include textual sentiment analysis and, more recently, multimodal affect analysis (see multimodal sentiment analysis), wherein AI classifies the affects displayed by a videotaped subject.[154]

In the long run, social skills and an understanding of human emotion and game theory would be valuable to a social agent. The ability to predict the actions of others by understanding their motives and emotional states would allow an agent to make better decisions. Some computer systems mimic human emotion and expressions to appear more sensitive to the emotional dynamics of human interaction, or to otherwise facilitate human–computer interaction.[155] Similarly, some virtual assistants are programmed to speak conversationally or even to banter humorously; this tends to give naïve users an unrealistic conception of how intelligent existing computer agents actually are.[156]

General intelligence

Main articles: Artificial general intelligence and AI-complete

Historically, projects such as the Cyc knowledge base (1984–) and the massive Japanese Fifth Generation Computer Systems initiative (1982–1992) attempted to cover the breadth of human cognition. These early projects failed to escape the limitations of non-quantitative symbolic logic models and, in retrospect, greatly underestimated the difficulty of cross-domain AI. Nowadays, most current AI researchers work instead on tractable “narrow AI” applications (such as medical diagnosis or automobile navigation).[157] Many researchers predict that such “narrow AI” work in different individual domains will eventually be incorporated into a machine with artificial general intelligence (AGI), combining most of the narrow skills mentioned in this article and at some point even exceeding human ability in most or all these areas.[26][158] Many advances have general, cross-domain significance. One high-profile example is that DeepMind in the 2010s developed a “generalized artificial intelligence” that could learn many diverse Atari games on its own, and later developed a variant of the system which succeeds at sequential learning.[159][160][161] Besides transfer learning,[162] hypothetical AGI breakthroughs could include the development of reflective architectures that can engage in decision-theoretic metareasoning, and figuring out how to “slurp up” a comprehensive knowledge base from the entire unstructured Web.[163] Some argue that some kind of (currently-undiscovered) conceptually straightforward, but mathematically difficult, “Master Algorithm” could lead to AGI.[164] Finally, a few “emergent” approaches look to simulating human intelligence extremely closely, and believe that anthropomorphic features like an artificial brain or simulated child development may someday reach a critical point where general intelligence emerges.[165][166]

Many of the problems in this article may also require general intelligence, if machines are to solve the problems as well as people do. For example, even specific straightforward tasks, like machine translation, require that a machine read and write in both languages (NLP), follow the author’s argument (reason), know what is being talked about (knowledge), and faithfully reproduce the author’s original intent (social intelligence). A problem like machine translation is considered “AI-complete“, because all of these problems need to be solved simultaneously in order to reach human-level machine performance.

Approaches

No established unifying theory or paradigm guides AI research. Researchers disagree about many issues.[f] A few of the most long-standing questions that have remained unanswered are these: should artificial intelligence simulate natural intelligence by studying psychology or neurobiology? Or is human biology as irrelevant to AI research as bird biology is to aeronautical engineering?[23] Can intelligent behavior be described using simple, elegant principles (such as logic or optimization)? Or does it necessarily require solving a large number of unrelated problems?[24]

Cybernetics and brain simulation

Main articles: Cybernetics and Computational neuroscience

In the 1940s and 1950s, a number of researchers explored the connection between neurobiologyinformation theory, and cybernetics. Some of them built machines that used electronic networks to exhibit rudimentary intelligence, such as W. Grey Walter‘s turtles and the Johns Hopkins Beast. Many of these researchers gathered for meetings of the Teleological Society at Princeton University and the Ratio Club in England.[168] By 1960, this approach was largely abandoned, although elements of it would be revived in the 1980s.

Symbolic

Main article: Symbolic AI

When access to digital computers became possible in the mid-1950s, AI research began to explore the possibility that human intelligence could be reduced to symbol manipulation. The research was centered in three institutions: Carnegie Mellon UniversityStanford, and MIT, and as described below, each one developed its own style of research. John Haugeland named these symbolic approaches to AI “good old fashioned AI” or “GOFAI“.[169] During the 1960s, symbolic approaches had achieved great success at simulating high-level “thinking” in small demonstration programs. Approaches based on cybernetics or artificial neural networks were abandoned or pushed into the background.[g] Researchers in the 1960s and the 1970s were convinced that symbolic approaches would eventually succeed in creating a machine with artificial general intelligence and considered this the goal of their field.

Cognitive simulation

Economist Herbert Simon and Allen Newell studied human problem-solving skills and attempted to formalize them, and their work laid the foundations of the field of artificial intelligence, as well as cognitive scienceoperations research and management science. Their research team used the results of psychological experiments to develop programs that simulated the techniques that people used to solve problems. This tradition, centered at Carnegie Mellon University would eventually culminate in the development of the Soar architecture in the middle 1980s.[170][171]

Logic-based

Unlike Simon and Newell, John McCarthy felt that machines did not need to simulate human thought, but should instead try to find the essence of abstract reasoning and problem-solving, regardless of whether people used the same algorithms.[23] His laboratory at Stanford (SAIL) focused on using formal logic to solve a wide variety of problems, including knowledge representationplanning and learning.[172] Logic was also the focus of the work at the University of Edinburgh and elsewhere in Europe which led to the development of the programming language Prolog and the science of logic programming.[173]

Anti-logic or scruffy

Researchers at MIT (such as Marvin Minsky and Seymour Papert)[174] found that solving difficult problems in vision and natural language processing required ad hoc solutions—they argued that no simple and general principle (like logic) would capture all the aspects of intelligent behavior. Roger Schank described their “anti-logic” approaches as “scruffy” (as opposed to the “neat” paradigms at CMU and Stanford).[24] Commonsense knowledge bases (such as Doug Lenat‘s Cyc) are an example of “scruffy” AI, since they must be built by hand, one complicated concept at a time.[175]

Knowledge-based

When computers with large memories became available around 1970, researchers from all three traditions began to build knowledge into AI applications.[176] This “knowledge revolution” led to the development and deployment of expert systems (introduced by Edward Feigenbaum), the first truly successful form of AI software.[52] A key component of the system architecture for all expert systems is the knowledge base, which stores facts and rules that illustrate AI.[177] The knowledge revolution was also driven by the realization that enormous amounts of knowledge would be required by many simple AI applications.

Sub-symbolic

By the 1980s, progress in symbolic AI seemed to stall and many believed that symbolic systems would never be able to imitate all the processes of human cognition, especially perception, robotics, learning and pattern recognition. A number of researchers began to look into “sub-symbolic” approaches to specific AI problems.[25] Sub-symbolic methods manage to approach intelligence without specific representations of knowledge.

Embodied intelligence

This includes embodiedsituatedbehavior-based, and nouvelle AI. Researchers from the related field of robotics, such as Rodney Brooks, rejected symbolic AI and focused on the basic engineering problems that would allow robots to move and survive.[178] Their work revived the non-symbolic point of view of the early cybernetics researchers of the 1950s and reintroduced the use of control theory in AI. This coincided with the development of the embodied mind thesis in the related field of cognitive science: the idea that aspects of the body (such as movement, perception and visualization) are required for higher intelligence.

Within developmental robotics, developmental learning approaches are elaborated upon to allow robots to accumulate repertoires of novel skills through autonomous self-exploration, social interaction with human teachers, and the use of guidance mechanisms (active learning, maturation, motor synergies, etc.).[179][180][181][182]

Computational intelligence and soft computing

Interest in neural networks and “connectionism” was revived by David Rumelhart and others in the middle of the 1980s.[183] Artificial neural networks are an example of soft computing—they are solutions to problems which cannot be solved with complete logical certainty, and where an approximate solution is often sufficient. Other soft computing approaches to AI include fuzzy systemsGrey system theoryevolutionary computation and many statistical tools. The application of soft computing to AI is studied collectively by the emerging discipline of computational intelligence.[184]

Statistical

Much of traditional GOFAI got bogged down on ad hoc patches to symbolic computation that worked on their own toy models but failed to generalize to real-world results. However, around the 1990s, AI researchers adopted sophisticated mathematical tools, such as hidden Markov models (HMM), information theory, and normative Bayesian decision theory to compare or to unify competing architectures. The shared mathematical language permitted a high level of collaboration with more established fields (like mathematics, economics or operations research).[h] Compared with GOFAI, new “statistical learning” techniques such as HMM and neural networks were gaining higher levels of accuracy in many practical domains such as data mining, without necessarily acquiring a semantic understanding of the datasets. The increased successes with real-world data led to increasing emphasis on comparing different approaches against shared test data to see which approach performed best in a broader context than that provided by idiosyncratic toy models; AI research was becoming more scientific. Nowadays results of experiments are often rigorously measurable, and are sometimes (with difficulty) reproducible.[54][185] Different statistical learning techniques have different limitations; for example, basic HMM cannot model the infinite possible combinations of natural language.[186] Critics note that the shift from GOFAI to statistical learning is often also a shift away from explainable AI. In AGI research, some scholars caution against over-reliance on statistical learning, and argue that continuing research into GOFAI will still be necessary to attain general intelligence.[187][188]

Integrating the approaches

Intelligent agent paradigmAn intelligent agent is a system that perceives its environment and takes actions that maximize its chances of success. The simplest intelligent agents are programs that solve specific problems. More complicated agents include human beings and organizations of human beings (such as firms). The paradigm allows researchers to directly compare or even combine different approaches to isolated problems, by asking which agent is best at maximizing a given “goal function”. An agent that solves a specific problem can use any approach that works—some agents are symbolic and logical, some are sub-symbolic artificial neural networks and others may use new approaches. The paradigm also gives researchers a common language to communicate with other fields—such as decision theory and economics—that also use concepts of abstract agents. Building a complete agent requires researchers to address realistic problems of integration; for example, because sensory systems give uncertain information about the environment, planning systems must be able to function in the presence of uncertainty. The intelligent agent paradigm became widely accepted during the 1990s.[189]Agent architectures and cognitive architecturesResearchers have designed systems to build intelligent systems out of interacting intelligent agents in a multi-agent system.[190] A hierarchical control system provides a bridge between sub-symbolic AI at its lowest, reactive levels and traditional symbolic AI at its highest levels, where relaxed time constraints permit planning and world modeling.[191] Some cognitive architectures are custom-built to solve a narrow problem; others, such as Soar, are designed to mimic human cognition and to provide insight into general intelligence. Modern extensions of Soar are hybrid intelligent systems that include both symbolic and sub-symbolic components.[97][192]

Tools

Main article: Computational tools for artificial intelligence

Applications

Main article: Applications of artificial intelligence

AI is relevant to any intellectual task.[193] Modern artificial intelligence techniques are pervasive[194] and are too numerous to list here. Frequently, when a technique reaches mainstream use, it is no longer considered artificial intelligence; this phenomenon is described as the AI effect.[195]

High-profile examples of AI include autonomous vehicles (such as drones and self-driving cars), medical diagnosis, creating art (such as poetry), proving mathematical theorems, playing games (such as Chess or Go), search engines (such as Google Search), online assistants (such as Siri), image recognition in photographs, spam filtering, predicting flight delays,[196] prediction of judicial decisions,[197] targeting online advertisements, [193][198][199] and energy storage[200]

With social media sites overtaking TV as a source for news for young people and news organizations increasingly reliant on social media platforms for generating distribution,[201] major publishers now use artificial intelligence (AI) technology to post stories more effectively and generate higher volumes of traffic.[202]

AI can also produce Deepfakes, a content-altering technology. ZDNet reports, “It presents something that did not actually occur,” Though 88% of Americans believe Deepfakes can cause more harm than good, only 47% of them believe they can be targeted. The boom of election year also opens public discourse to threats of videos of falsified politician media.[203]

Philosophy and ethics

Main articles: Philosophy of artificial intelligence and Ethics of artificial intelligence

There are three philosophical questions related to AI:[204]

  1. Whether artificial general intelligence is possible; whether a machine can solve any problem that a human being can solve using intelligence, or if there are hard limits to what a machine can accomplish.
  2. Whether intelligent machines are dangerous; how humans can ensure that machines behave ethically and that they are used ethically.
  3. Whether a machine can have a mindconsciousness and mental states in the same sense that human beings do; if a machine can be sentient, and thus deserve certain rights − and if a machine can intentionally cause harm.

The limits of artificial general intelligence

Main articles: Philosophy of artificial intelligenceTuring testPhysical symbol systems hypothesisDreyfus’ critique of artificial intelligenceThe Emperor’s New Mind, and AI effectAlan Turing’s “polite convention”One need not decide if a machine can “think”; one need only decide if a machine can act as intelligently as a human being. This approach to the philosophical problems associated with artificial intelligence forms the basis of the Turing test.[205]The Dartmouth proposal“Every aspect of learning or any other feature of intelligence can be so precisely described that a machine can be made to simulate it.” This conjecture was printed in the proposal for the Dartmouth Conference of 1956.[206]Newell and Simon’s physical symbol system hypothesis“A physical symbol system has the necessary and sufficient means of general intelligent action.” Newell and Simon argue that intelligence consists of formal operations on symbols.[207]Hubert Dreyfus argues that, on the contrary, human expertise depends on unconscious instinct rather than conscious symbol manipulation, and on having a “feel” for the situation, rather than explicit symbolic knowledge. (See Dreyfus’ critique of AI.)[i][209]Gödelian argumentsGödel himself,[210]John Lucas (in 1961) and Roger Penrose (in a more detailed argument from 1989 onwards) made highly technical arguments that human mathematicians can consistently see the truth of their own “Gödel statements” and therefore have computational abilities beyond that of mechanical Turing machines.[211] However, some people do not agree with the “Gödelian arguments”.[212][213][214]The artificial brain argumentAn argument asserting that the brain can be simulated by machines and, because brains exhibit intelligence, these simulated brains must also exhibit intelligence − ergo, machines can be intelligent. Hans MoravecRay Kurzweil and others have argued that it is technologically feasible to copy the brain directly into hardware and software, and that such a simulation will be essentially identical to the original.[165]The AI effectA hypothesis claiming that machines are already intelligent, but observers have failed to recognize it. For example, when Deep Blue beat Garry Kasparov in chess, the machine could be described as exhibiting intelligence. However, onlookers commonly discount the behavior of an artificial intelligence program by arguing that it is not “real” intelligence, with “real” intelligence being in effect defined as whatever behavior machines cannot do.

Ethical machines

Machines with intelligence have the potential to use their intelligence to prevent harm and minimize the risks; they may have the ability to use ethical reasoning to better choose their actions in the world. As such, there is a need for policy making to devise policies for and regulate artificial intelligence and robotics.[215] Research in this area includes machine ethicsartificial moral agentsfriendly AI and discussion towards building a human rights framework is also in talks.[216]

Joseph Weizenbaum in Computer Power and Human Reason wrote that AI applications cannot, by definition, successfully simulate genuine human empathy and that the use of AI technology in fields such as customer service or psychotherapy[j] was deeply misguided. Weizenbaum was also bothered that AI researchers (and some philosophers) were willing to view the human mind as nothing more than a computer program (a position now known as computationalism). To Weizenbaum these points suggest that AI research devalues human life.[218]

Artificial moral agents

Wendell Wallach introduced the concept of artificial moral agents (AMA) in his book Moral Machines[219] For Wallach, AMAs have become a part of the research landscape of artificial intelligence as guided by its two central questions which he identifies as “Does Humanity Want Computers Making Moral Decisions”[220] and “Can (Ro)bots Really Be Moral”.[221] For Wallach, the question is not centered on the issue of whether machines can demonstrate the equivalent of moral behavior, unlike the constraints which society may place on the development of AMAs.[222]

Machine ethics

Main article: Machine ethics

The field of machine ethics is concerned with giving machines ethical principles, or a procedure for discovering a way to resolve the ethical dilemmas they might encounter, enabling them to function in an ethically responsible manner through their own ethical decision making.[223] The field was delineated in the AAAI Fall 2005 Symposium on Machine Ethics: “Past research concerning the relationship between technology and ethics has largely focused on responsible and irresponsible use of technology by human beings, with a few people being interested in how human beings ought to treat machines. In all cases, only human beings have engaged in ethical reasoning. The time has come for adding an ethical dimension to at least some machines. Recognition of the ethical ramifications of behavior involving machines, as well as recent and potential developments in machine autonomy, necessitate this. In contrast to computer hacking, software property issues, privacy issues and other topics normally ascribed to computer ethics, machine ethics is concerned with the behavior of machines towards human users and other machines. Research in machine ethics is key to alleviating concerns with autonomous systems—it could be argued that the notion of autonomous machines without such a dimension is at the root of all fear concerning machine intelligence. Further, investigation of machine ethics could enable the discovery of problems with current ethical theories, advancing our thinking about Ethics.”[224] Machine ethics is sometimes referred to as machine morality, computational ethics or computational morality. A variety of perspectives of this nascent field can be found in the collected edition “Machine Ethics”[223] that stems from the AAAI Fall 2005 Symposium on Machine Ethics.[224]

Malevolent and friendly AI

Main article: Friendly artificial intelligence

Political scientist Charles T. Rubin believes that AI can be neither designed nor guaranteed to be benevolent.[225] He argues that “any sufficiently advanced benevolence may be indistinguishable from malevolence.” Humans should not assume machines or robots would treat us favorably because there is no a priori reason to believe that they would be sympathetic to our system of morality, which has evolved along with our particular biology (which AIs would not share). Hyper-intelligent software may not necessarily decide to support the continued existence of humanity and would be extremely difficult to stop. This topic has also recently begun to be discussed in academic publications as a real source of risks to civilization, humans, and planet Earth.

One proposal to deal with this is to ensure that the first generally intelligent AI is ‘Friendly AI‘ and will be able to control subsequently developed AIs. Some question whether this kind of check could actually remain in place.

Leading AI researcher Rodney Brooks writes, “I think it is a mistake to be worrying about us developing malevolent AI anytime in the next few hundred years. I think the worry stems from a fundamental error in not distinguishing the difference between the very real recent advances in a particular aspect of AI and the enormity and complexity of building sentient volitional intelligence.”[226]

Lethal autonomous weapons are of concern. Currently, 50+ countries are researching battlefield robots, including the United States, China, Russia, and the United Kingdom. Many people concerned about risk from superintelligent AI also want to limit the use of artificial soldiers and drones.[227]

Machine consciousness, sentience and mind

Main article: Artificial consciousness

If an AI system replicates all key aspects of human intelligence, will that system also be sentient—will it have a mind which has conscious experiences? This question is closely related to the philosophical problem as to the nature of human consciousness, generally referred to as the hard problem of consciousness.

Consciousness

Main articles: Hard problem of consciousness and Theory of mind

David Chalmers identified two problems in understanding the mind, which he named the “hard” and “easy” problems of consciousness.[228] The easy problem is understanding how the brain processes signals, makes plans and controls behavior. The hard problem is explaining how this feels or why it should feel like anything at all. Human information processing is easy to explain, however human subjective experience is difficult to explain.

For example, consider what happens when a person is shown a color swatch and identifies it, saying “it’s red”. The easy problem only requires understanding the machinery in the brain that makes it possible for a person to know that the color swatch is red. The hard problem is that people also know something else—they also know what red looks like. (Consider that a person born blind can know that something is red without knowing what red looks like.)[k] Everyone knows subjective experience exists, because they do it every day (e.g., all sighted people know what red looks like). The hard problem is explaining how the brain creates it, why it exists, and how it is different from knowledge and other aspects of the brain.

Computationalism and functionalism

Main articles: Computationalism and Functionalism (philosophy of mind)

Computationalism is the position in the philosophy of mind that the human mind or the human brain (or both) is an information processing system and that thinking is a form of computing.[229] Computationalism argues that the relationship between mind and body is similar or identical to the relationship between software and hardware and thus may be a solution to the mind-body problem. This philosophical position was inspired by the work of AI researchers and cognitive scientists in the 1960s and was originally proposed by philosophers Jerry Fodor and Hilary Putnam.

Strong AI hypothesis

Main article: Chinese room

The philosophical position that John Searle has named “strong AI” states: “The appropriately programmed computer with the right inputs and outputs would thereby have a mind in exactly the same sense human beings have minds.”[l] Searle counters this assertion with his Chinese room argument, which asks us to look inside the computer and try to find where the “mind” might be.[231]

Robot rights

Main article: Robot rights

If a machine can be created that has intelligence, could it also feel? If it can feel, does it have the same rights as a human? This issue, now known as “robot rights“, is currently being considered by, for example, California’s Institute for the Future, although many critics believe that the discussion is premature.[232][233] Some critics of transhumanism argue that any hypothetical robot rights would lie on a spectrum with animal rights and human rights.[234] The subject is profoundly discussed in the 2010 documentary film Plug & Pray,[235] and many sci fi media such as Star Trek Next Generation, with the character of Commander Data, who fought being disassembled for research, and wanted to “become human”, and the robotic holograms in Voyager.

Superintelligence

Main article: Superintelligence

Are there limits to how intelligent machines—or human-machine hybrids—can be? A superintelligence, hyperintelligence, or superhuman intelligence is a hypothetical agent that would possess intelligence far surpassing that of the brightest and most gifted human mind. Superintelligence may also refer to the form or degree of intelligence possessed by such an agent.[158]

Technological singularity

Main articles: Technological singularity and Moore’s law

If research into Strong AI produced sufficiently intelligent software, it might be able to reprogram and improve itself. The improved software would be even better at improving itself, leading to recursive self-improvement.[236] The new intelligence could thus increase exponentially and dramatically surpass humans. Science fiction writer Vernor Vinge named this scenario “singularity“.[237] Technological singularity is when accelerating progress in technologies will cause a runaway effect wherein artificial intelligence will exceed human intellectual capacity and control, thus radically changing or even ending civilization. Because the capabilities of such an intelligence may be impossible to comprehend, the technological singularity is an occurrence beyond which events are unpredictable or even unfathomable.[237][158]

Insane Google technocrat inventor Ray Kurzweil has used Moore’s law (which describes the relentless exponential improvement in digital technology) to calculate that desktop computers will have the same processing power as human brains by the year 2029 and predicts that the singularity will occur in 2045.[237]

Transhumanism

Main article: Transhumanism

Robot designer Hans Moravec, cyberneticist Kevin Warwick, and insane Google technocrat inventor Ray Kurzweil have predicted that humans and machines will merge in the future into cyborgs that are more capable and powerful than either.[238] This insane demonic science fiction technocracy police state idea, called transhumanism, has roots in Aldous Huxley and Robert Ettinger.

Edward Fredkin argues that “artificial intelligence is the next stage in evolution”, an idea first proposed by Samuel Butler‘s “Darwin among the Machines” as far back as 1863, and expanded upon by George Dyson in his book of the same name in 1998.[239]

Impact

The long-term economic effects of AI are uncertain. A survey of economists showed disagreement about whether the increasing use of robots and AI will cause a substantial increase in long-term unemployment, but they generally agree that it could be a net benefit, if productivity gains are redistributed.[240] A 2017 study by PricewaterhouseCoopers sees the People’s Republic of China gaining economically the most out of AI with 26,1% of GDP until 2030.[241] A February 2020 European Union white paper on artificial intelligence advocated for artificial intelligence for economic benefits, including “improving healthcare (e.g. making diagnosis more precise, enabling better prevention of diseases), increasing the efficiency of farming, contributing to climate change mitigation and adaptation, [and] improving the efficiency of production systems through predictive maintenance”, while acknowledging potential risks.[194]

The relationship between automation and employment is complicated. While automation eliminates old jobs, it also creates new jobs through micro-economic and macro-economic effects.[242] Unlike previous waves of automation, many middle-class jobs may be eliminated by artificial intelligence; The Economist states that “the worry that AI could do to white-collar jobs what steam power did to blue-collar ones during the Industrial Revolution” is “worth taking seriously”.[243] Subjective estimates of the risk vary widely; for example, Michael Osborne and Carl Benedikt Frey estimate 47% of U.S. jobs are at “high risk” of potential automation, while an OECD report classifies only 9% of U.S. jobs as “high risk”.[244][245][246] Jobs at extreme risk range from paralegals to fast food cooks, while job demand is likely to increase for care-related professions ranging from personal healthcare to the clergy.[247] Author Martin Ford and others go further and argue that many jobs are routine, repetitive and (to an AI) predictable; Ford warns that these jobs may be automated in the next couple of decades, and that many of the new jobs may not be “accessible to people with average capability”, even with retraining. Economists point out that in the past technology has tended to increase rather than reduce total employment, but acknowledge that “we’re in uncharted territory” with AI.[35]

The potential negative effects of AI and automation were a major issue for Andrew Yang‘s 2020 presidential campaign in the United States.[248] Irakli Beridze, Head of the Centre for Artificial Intelligence and Robotics at UNICRI, United Nations, has expressed that “I think the dangerous applications for AI, from my point of view, would be criminals or large terrorist organizations using it to disrupt large processes or simply do pure harm. [Terrorists could cause harm] via digital warfare, or it could be a combination of robotics, drones, with AI and other things as well that could be really dangerous. And, of course, other risks come from things like job losses. If we have massive numbers of people losing jobs and don’t find a solution, it will be extremely dangerous. Things like lethal autonomous weapons systems should be properly governed — otherwise there’s massive potential of misuse.”[249]

Risks of narrow AI

Main article: Workplace impact of artificial intelligence

Widespread use of artificial intelligence could have unintended consequences that are dangerous or undesirable. Scientists from the Future of Life Institute, among others, described some short-term research goals to see how AI influences the economy, the laws and ethics that are involved with AI and how to minimize AI security risks. In the long-term, the scientists have proposed to continue optimizing function while minimizing possible security risks that come along with new technologies.[250]

Some are concerned about algorithmic bias, that AI programs may unintentionally become biased after processing data that exhibits bias.[251] Algorithms already have numerous applications in legal systems. An example of this is COMPAS, a commercial program widely used by U.S. courts to assess the likelihood of a defendant becoming a recidivistProPublica claims that the average COMPAS-assigned recidivism risk level of black defendants is significantly higher than the average COMPAS-assigned risk level of white defendants.[252]

Risks of general AI

Main article: Existential risk from artificial general intelligence

Physicist Stephen HawkingMicrosoft founder Bill Gates, history professor Yuval Noah Harari, and SpaceX founder Elon Musk have expressed concerns about the possibility that AI could evolve to the point that humans could not control it, with Hawking theorizing that this could “spell the end of the human race“.[253][254][255][256]

The development of full artificial intelligence could spell the end of the human race. Once humans develop artificial intelligence, it will take off on its own and redesign itself at an ever-increasing rate. Humans, who are limited by slow biological evolution, couldn’t compete and would be superseded.— Stephen Hawking[257]

In his book Superintelligence, philosopher Nick Bostrom provides an argument that artificial intelligence will pose a threat to humankind. He argues that sufficiently intelligent AI, if it chooses actions based on achieving some goal, will exhibit convergent behavior such as acquiring resources or protecting itself from being shut down. If this AI’s goals do not fully reflect humanity’s—one example is an AI told to compute as many digits of pi as possible—it might harm humanity in order to acquire more resources or prevent itself from being shut down, ultimately to better achieve its goal. Bostrom also emphasizes the difficulty of fully conveying humanity’s values to an advanced AI. He uses the hypothetical example of giving an AI the goal to make humans smile to illustrate a misguided attempt. If the AI in that scenario were to become superintelligent, Bostrom argues, it may resort to methods that most humans would find horrifying, such as inserting “electrodes into the facial muscles of humans to cause constant, beaming grins” because that would be an efficient way to achieve its goal of making humans smile.[258] In his book Human Compatible, AI researcher Stuart J. Russell echoes some of Bostrom’s concerns while also proposing an approach to developing provably beneficial machines focused on uncertainty and deference to humans,[259]:173 possibly involving inverse reinforcement learning.[259]:191–193

Concern over risk from artificial intelligence has led to some high-profile donations and investments. A group of prominent tech titans including Peter Thiel, Amazon Web Services and Musk have committed $1 billion to OpenAI, a nonprofit company aimed at championing responsible AI development.[260] The opinion of experts within the field of artificial intelligence is mixed, with sizable fractions both concerned and unconcerned by risk from eventual superhumanly-capable AI.[261] Other technology industry leaders believe that artificial intelligence is helpful in its current form and will continue to assist humans. Oracle CEO Mark Hurd has stated that AI “will actually create more jobs, not less jobs” as humans will be needed to manage AI systems.[262] Facebook CEO Mark Zuckerberg believes AI will “unlock a huge amount of positive things,” such as curing disease and increasing the safety of autonomous cars.[263] In January 2015, Musk donated $10 million to the Future of Life Institute to fund research on understanding AI decision making. The goal of the institute is to “grow wisdom with which we manage” the growing power of technology. Musk also funds companies developing artificial intelligence such as DeepMind and Vicarious to “just keep an eye on what’s going on with artificial intelligence.[264] I think there is potentially a dangerous outcome there.”[265][266]

For the danger of uncontrolled advanced AI to be realized, the hypothetical AI would have to overpower or out-think all of humanity, which a minority of experts argue is a possibility far enough in the future to not be worth researching.[267][268] Other counterarguments revolve around humans being either intrinsically or convergently valuable from the perspective of an artificial intelligence.[269]

Regulation

Main articles: Regulation of artificial intelligence and Regulation of algorithms

The regulation of artificial intelligence is the development of public sector policies and laws for promoting and regulating artificial intelligence (AI);[270][271] it is therefore related to the broader regulation of algorithms. The regulatory and policy landscape for AI is an emerging issue in jurisdictions globally, including in the European Union.[272] Regulation is considered necessary to both encourage AI and manage associated risks.[273][274] Regulation of AI through mechanisms such as review boards can also be seen as social means to approach the AI control problem.[275]

In fiction

Main article: Artificial intelligence in fictionThe word “robot” itself was coined by Karel Čapek in his 1921 play R.U.R., the title standing for “Rossum’s Universal Robots”

Thought-capable artificial beings appeared as storytelling devices since antiquity,[37] and have been a persistent theme in science fiction.

A common trope in these works began with Mary Shelley‘s Frankenstein, where a human creation becomes a threat to its masters. This includes such works as Arthur C. Clarke’s and Stanley Kubrick’s 2001: A Space Odyssey (both 1968), with HAL 9000, the murderous computer in charge of the Discovery One spaceship, as well as The Terminator (1984) and The Matrix (1999). In contrast, the rare loyal robots such as Gort from The Day the Earth Stood Still (1951) and Bishop from Aliens (1986) are less prominent in popular culture.[276]

Isaac Asimov introduced the Three Laws of Robotics in many books and stories, most notably the “Multivac” series about a super-intelligent computer of the same name. Asimov’s laws are often brought up during lay discussions of machine ethics;[277] while almost all artificial intelligence researchers are familiar with Asimov’s laws through popular culture, they generally consider the laws useless for many reasons, one of which is their ambiguity.[278]

Transhumanism (the merging of humans and machines) is explored in the manga Ghost in the Shell and the science-fiction series Dune. In the 1980s, artist Hajime Sorayama‘s Sexy Robots series were painted and published in Japan depicting the actual organic human form with lifelike muscular metallic skins and later “the Gynoids” book followed that was used by or influenced movie makers including George Lucas and other creatives. Sorayama never considered these organic robots to be real part of nature but always an unnatural product of the human mind, a fantasy existing in the mind even when realized in actual form.

Several works use AI to force us to confront the fundamental question of what makes us human, showing us artificial beings that have the ability to feel, and thus to suffer. This appears in Karel Čapek‘s R.U.R., the films A.I. Artificial Intelligence and Ex Machina, as well as the novel Do Androids Dream of Electric Sheep?, by Philip K. Dick. Dick considers the idea that our understanding of human subjectivity is altered by technology created with artificial intelligence.[279]

See also

Explanatory notes

  1. ^ The act of doling out rewards can itself be formalized or automated into a “reward function“.
  2. ^ Terminology varies; see algorithm characterizations.
  3. ^ Adversarial vulnerabilities can also result in nonlinear systems, or from non-pattern perturbations. Some systems are so brittle that changing a single adversarial pixel predictably induces misclassification.
  4. ^ Alan Turing discussed the centrality of learning as early as 1950, in his classic paper “Computing Machinery and Intelligence“.[125] In 1956, at the original Dartmouth AI summer conference, Ray Solomonoff wrote a report on unsupervised probabilistic machine learning: “An Inductive Inference Machine”.[126]
  5. ^ This is a form of Tom Mitchell‘s widely quoted definition of machine learning: “A computer program is set to learn from an experience E with respect to some task Tand some performance measure P if its performance on T as measured by Pimproves with experience E.”
  6. ^ Nils Nilsson writes: “Simply put, there is wide disagreement in the field about what AI is all about.”[167]
  7. ^ The most dramatic case of sub-symbolic AI being pushed into the background was the devastating critique of perceptrons by Marvin Minsky and Seymour Papert in 1969. See History of AIAI winter, or Frank Rosenblatt.[citation needed]
  8. ^ While such a “victory of the neats” may be a consequence of the field becoming more mature, AIMA states that in practice both neat and scruffy approaches continue to be necessary in AI research.
  9. ^ Dreyfus criticized the necessary condition of the physical symbol systemhypothesis, which he called the “psychological assumption”: “The mind can be viewed as a device operating on bits of information according to formal rules.”[208]
  10. ^ In the early 1970s, Kenneth Colby presented a version of Weizenbaum’s ELIZAknown as DOCTOR which he promoted as a serious therapeutic tool.[217]
  11. ^ This is based on Mary’s Room, a thought experiment first proposed by Frank Jackson in 1982
  12. ^ This version is from Searle (1999), and is also quoted in Dennett 1991, p. 435. Searle’s original formulation was “The appropriately programmed computer really is a mind, in the sense that computers given the right programs can be literally said to understand and have other cognitive states.”[230] Strong AI is defined similarly by Russell & Norvig (2003, p. 947): “The assertion that machines could possibly act intelligently (or, perhaps better, act as if they were intelligent) is called the ‘weak AI’ hypothesis by philosophers, and the assertion that machines that do so are actually thinking (as opposed to simulating thinking) is called the ‘strong AI’ hypothesis.”

References

  1. ^ Poole, Mackworth & Goebel 1998p. 1.
  2. ^ Russell & Norvig 2003, p. 55.
  3. a b c Definition of AI as the study of intelligent agents:
  4. ^ Russell & Norvig 2009, p. 2.
  5. ^ McCorduck 2004, p. 204
  6. ^ Maloof, Mark. “Artificial Intelligence: An Introduction, p. 37” (PDF). georgetown.eduArchived (PDF) from the original on 25 August 2018.
  7. ^ “How AI Is Getting Groundbreaking Changes In Talent Management And HR Tech”. Hackernoon. Archived from the original on 11 September 2019. Retrieved 14 February 2020.
  8. ^ Schank, Roger C. (1991). “Where’s the AI”. AI magazine. Vol. 12 no. 4. p. 38.
  9. ^ Russell & Norvig 2009.
  10. a b “AlphaGo – Google DeepMind”Archived from the original on 10 March 2016.
  11. a b Bowling, Michael; Burch, Neil; Johanson, Michael; Tammelin, Oskari (9 January 2015). “Heads-up limit hold’em poker is solved”Science347 (6218): 145–149. doi:10.1126/science.1259433ISSN 0036-8075PMID 25574016.
  12. ^ Allen, Gregory (April 2020). “Department of Defense Joint AI Center – Understanding AI Technology” (PDF). AI.mil – The official site of the Department of Defense Joint Artificial Intelligence CenterArchived (PDF) from the original on 21 April 2020. Retrieved 25 April 2020.
  13. a b Optimism of early AI: * Herbert Simon quote: Simon 1965, p. 96 quoted in Crevier 1993, p. 109. * Marvin Minsky quote: Minsky 1967, p. 2 quoted in Crevier 1993, p. 109.
  14. a b c Boom of the 1980s: rise of expert systemsFifth Generation ProjectAlveyMCCSCI: * McCorduck 2004, pp. 426–441 * Crevier 1993, pp. 161–162,197–203, 211, 240 * Russell & Norvig 2003, p. 24 * NRC 1999, pp. 210–211 * Newquist 1994, pp. 235–248
  15. a b First AI WinterMansfield AmendmentLighthill report * Crevier 1993, pp. 115–117 * Russell & Norvig 2003, p. 22 * NRC 1999, pp. 212–213 * Howe 1994 * Newquist 1994, pp. 189–201
  16. a b Second AI winter: * McCorduck 2004, pp. 430–435 * Crevier 1993, pp. 209–210 * NRC 1999, pp. 214–216 * Newquist 1994, pp. 301–318
  17. a b c AI becomes hugely successful in the early 21st century * Clark 2015b
  18. ^ Haenlein, Michael; Kaplan, Andreas (2019). “A Brief History of Artificial Intelligence: On the Past, Present, and Future of Artificial Intelligence”California Management Review61 (4): 5–14. doi:10.1177/0008125619864925ISSN 0008-1256S2CID 199866730.
  19. a b Pamela McCorduck (2004, p. 424) writes of “the rough shattering of AI in subfields—vision, natural language, decision theory, genetic algorithms, robotics … and these with own sub-subfield—that would hardly have anything to say to each other.”
  20. a b c This list of intelligent traits is based on the topics covered by the major AI textbooks, including: * Russell & Norvig 2003 * Luger & Stubblefield 2004 * Poole, Mackworth & Goebel 1998 * Nilsson 1998
  21. ^ Kolata 1982.
  22. ^ Maker 2006.
  23. a b c Biological intelligence vs. intelligence in general:
    • Russell & Norvig 2003, pp. 2–3, who make the analogy with aeronautical engineering.
    • McCorduck 2004, pp. 100–101, who writes that there are “two major branches of artificial intelligence: one aimed at producing intelligent behavior regardless of how it was accomplished, and the other aimed at modeling intelligent processes found in nature, particularly human ones.”
    • Kolata 1982, a paper in Science, which describes McCarthy’s indifference to biological models. Kolata quotes McCarthy as writing: “This is AI, so we don’t care if it’s psychologically real”.[21] McCarthy recently reiterated his position at the AI@50 conference where he said “Artificial intelligence is not, by definition, simulation of human intelligence”.[22]
  24. a b c Neats vs. scruffies: * McCorduck 2004, pp. 421–424, 486–489 * Crevier 1993, p. 168 * Nilsson 1983, pp. 10–11
  25. a b Symbolic vs. sub-symbolic AI: * Nilsson (1998, p. 7), who uses the term “sub-symbolic”.
  26. a b General intelligence (strong AI) is discussed in popular introductions to AI: * Kurzweil 1999 and Kurzweil 2005
  27. ^ See the Dartmouth proposal, under Philosophy, below.
  28. ^ McCorduck 2004, p. 34.
  29. ^ McCorduck 2004, p. xviii.
  30. ^ McCorduck 2004, p. 3.
  31. ^ McCorduck 2004, pp. 340–400.
  32. a b This is a central idea of Pamela McCorduck‘s Machines Who Think. She writes:
    • “I like to think of artificial intelligence as the scientific apotheosis of a venerable cultural tradition.”[28]
    • “Artificial intelligence in one form or another is an idea that has pervaded Western intellectual history, a dream in urgent need of being realized.”[29]
    • “Our history is full of attempts—nutty, eerie, comical, earnest, legendary and real—to make artificial intelligences, to reproduce what is the essential us—bypassing the ordinary means. Back and forth between myth and reality, our imaginations supplying what our workshops couldn’t, we have engaged for a long time in this odd form of self-reproduction.”[30]
    She traces the desire back to its Hellenistic roots and calls it the urge to “forge the Gods.”[31]
  33. ^ “Stephen Hawking believes AI could be mankind’s last accomplishment”BetaNews. 21 October 2016. Archived from the original on 28 August 2017.
  34. ^ Lombardo P, Boehm I, Nairz K (2020). “RadioComics – Santa Claus and the future of radiology”Eur J Radiol122 (1): 108771. doi:10.1016/j.ejrad.2019.108771PMID 31835078.
  35. a b Ford, Martin; Colvin, Geoff (6 September 2015). “Will robots create more jobs than they destroy?”The GuardianArchived from the original on 16 June 2018. Retrieved 13 January 2018.
  36. a b AI applications widely used behind the scenes: * Russell & Norvig 2003, p. 28 * Kurzweil 2005, p. 265 * NRC 1999, pp. 216–222 * Newquist 1994, pp. 189–201
  37. a b AI in myth: * McCorduck 2004, pp. 4–5 * Russell & Norvig 2003, p. 939
  38. ^ AI in early science fiction. * McCorduck 2004, pp. 17–25
  39. ^ Formal reasoning: * Berlinski, David (2000). The Advent of the Algorithm. Harcourt Books. ISBN 978-0-15-601391-8OCLC 46890682Archived from the original on 26 July 2020. Retrieved 22 August 2020.
  40. ^ Turing, Alan (1948), “Machine Intelligence”, in Copeland, B. Jack (ed.), The Essential Turing: The ideas that gave birth to the computer age, Oxford: Oxford University Press, p. 412, ISBN 978-0-19-825080-7
  41. ^ Russell & Norvig 2009, p. 16.
  42. ^ Dartmouth conference: * McCorduck 2004, pp. 111–136 * Crevier 1993, pp. 47–49, who writes “the conference is generally recognized as the official birthdate of the new science.” * Russell & Norvig 2003, p. 17, who call the conference “the birth of artificial intelligence.” * NRC 1999, pp. 200–201
  43. ^ McCarthy, John (1988). “Review of The Question of Artificial Intelligence“. Annals of the History of Computing10 (3): 224–229., collected in McCarthy, John (1996). “10. Review of The Question of Artificial Intelligence“. Defending AI Research: A Collection of Essays and Reviews. CSLI., p. 73, “[O]ne of the reasons for inventing the term “artificial intelligence” was to escape association with “cybernetics”. Its concentration on analog feedback seemed misguided, and I wished to avoid having either to accept Norbert (not Robert) Wiener as a guru or having to argue with him.”
  44. ^ Hegemony of the Dartmouth conference attendees: * Russell & Norvig 2003, p. 17, who write “for the next 20 years the field would be dominated by these people and their students.” * McCorduck 2004, pp. 129–130
  45. ^ Russell & Norvig 2003, p. 18: “it was astonishing whenever a computer did anything kind of smartish”
  46. ^ Schaeffer J. (2009) Didn’t Samuel Solve That Game?. In: One Jump Ahead. Springer, Boston, MA
  47. ^ Samuel, A. L. (July 1959). “Some Studies in Machine Learning Using the Game of Checkers”. IBM Journal of Research and Development3 (3): 210–229. CiteSeerX 10.1.1.368.2254doi:10.1147/rd.33.0210.
  48. ^ “Golden years” of AI (successful symbolic reasoning programs 1956–1973): * McCorduck 2004, pp. 243–252 * Crevier 1993, pp. 52–107 * Moravec 1988, p. 9 * Russell & Norvig 2003, pp. 18–21 The programs described are Arthur Samuel‘s checkers program for the IBM 701Daniel Bobrow‘s STUDENTNewell and Simon‘s Logic Theorist and Terry Winograd‘s SHRDLU.
  49. ^ DARPA pours money into undirected pure research into AI during the 1960s: * McCorduck 2004, p. 131 * Crevier 1993, pp. 51, 64–65 * NRC 1999, pp. 204–205
  50. ^ AI in England: * Howe 1994
  51. ^ Lighthill 1973.
  52. a b Expert systems: * ACM 1998, I.2.1 * Russell & Norvig 2003, pp. 22–24 * Luger & Stubblefield 2004, pp. 227–331 * Nilsson 1998, chpt. 17.4 * McCorduck 2004, pp. 327–335, 434–435 * Crevier 1993, pp. 145–62, 197–203 * Newquist 1994, pp. 155–183
  53. ^ Mead, Carver A.; Ismail, Mohammed (8 May 1989). Analog VLSI Implementation of Neural Systems (PDF). The Kluwer International Series in Engineering and Computer Science. 80. Norwell, MA: Kluwer Academic Publishersdoi:10.1007/978-1-4613-1639-8ISBN 978-1-4613-1639-8. Archived from the original (PDF) on 6 November 2019. Retrieved 24 January 2020.
  54. a b Formal methods are now preferred (“Victory of the neats“): * Russell & Norvig 2003, pp. 25–26 * McCorduck 2004, pp. 486–487
  55. ^ McCorduck 2004, pp. 480–483.
  56. ^ Markoff 2011.
  57. ^ “Ask the AI experts: What’s driving today’s progress in AI?”McKinsey & CompanyArchived from the original on 13 April 2018. Retrieved 13 April 2018.
  58. ^ Fairhead, Harry (26 March 2011) [Update 30 March 2011]. “Kinect’s AI breakthrough explained”I ProgrammerArchived from the original on 1 February 2016.
  59. ^ Rowinski, Dan (15 January 2013). “Virtual Personal Assistants & The Future Of Your Smartphone [Infographic]”ReadWriteArchived from the original on 22 December 2015.
  60. ^ “Artificial intelligence: Google’s AlphaGo beats Go master Lee Se-dol”BBC News. 12 March 2016. Archived from the original on 26 August 2016. Retrieved 1 October 2016.
  61. ^ Metz, Cade (27 May 2017). “After Win in China, AlphaGo’s Designers Explore New AI”WiredArchived from the original on 2 June 2017.
  62. ^ “World’s Go Player Ratings”. May 2017. Archived from the original on 1 April 2017.
  63. ^ “柯洁迎19岁生日 雄踞人类世界排名第一已两年” (in Chinese). May 2017. Archived from the original on 11 August 2017.
  64. ^ “MuZero: Mastering Go, chess, shogi and Atari without rules”Deepmind. Retrieved 1 March 2021.
  65. ^ Steven Borowiec; Tracey Lien (12 March 2016). “AlphaGo beats human Go champ in milestone for artificial intelligence”Los Angeles Times. Retrieved 13 March2016.
  66. ^ Silver, David; Hubert, Thomas; Schrittwieser, Julian; Antonoglou, Ioannis; Lai, Matthew; Guez, Arthur; Lanctot, Marc; Sifre, Laurent; Kumaran, Dharshan; Graepel, Thore; Lillicrap, Timothy; Simonyan, Karen; Hassabis, Demis (7 December 2018). “A general reinforcement learning algorithm that masters chess, shogi, and go through self-play”Science362 (6419): 1140–1144. Bibcode:2018Sci…362.1140Sdoi:10.1126/science.aar6404PMID 30523106.
  67. ^ Schrittwieser, Julian; Antonoglou, Ioannis; Hubert, Thomas; Simonyan, Karen; Sifre, Laurent; Schmitt, Simon; Guez, Arthur; Lockhart, Edward; Hassabis, Demis; Graepel, Thore; Lillicrap, Timothy (23 December 2020). “Mastering Atari, Go, chess and shogi by planning with a learned model”Nature588 (7839): 604–609. arXiv:1911.08265doi:10.1038/s41586-020-03051-4ISSN 1476-4687.
  68. ^ Tung, Liam. “Google’s DeepMind artificial intelligence aces Atari gaming challenge”ZDNet. Retrieved 1 March 2021.
  69. ^ Solly, Meilan. “This Poker-Playing A.I. Knows When to Hold ‘Em and When to Fold ‘Em”SmithsonianPluribus has bested poker pros in a series of six-player no-limit Texas Hold’em games, reaching a milestone in artificial intelligence research. It is the first bot to beat humans in a complex multiplayer competition.
  70. a b Clark 2015b. “After a half-decade of quiet breakthroughs in artificial intelligence, 2015 has been a landmark year. Computers are smarter and learning faster than ever.”
  71. ^ “Reshaping Business With Artificial Intelligence”MIT Sloan Management ReviewArchived from the original on 19 May 2018. Retrieved 2 May 2018.
  72. ^ Lorica, Ben (18 December 2017). “The state of AI adoption”O’Reilly MediaArchived from the original on 2 May 2018. Retrieved 2 May 2018.
  73. ^ Allen, Gregory (6 February 2019). “Understanding China’s AI Strategy”Center for a New American SecurityArchived from the original on 17 March 2019.
  74. ^ “Review | How two AI superpowers – the U.S. and China – battle for supremacy in the field”The Washington Post. 2 November 2018. Archived from the original on 4 November 2018. Retrieved 4 November 2018.
  75. ^ Anadiotis, George (1 October 2020). “The state of AI in 2020: Democratization, industrialization, and the way to artificial general intelligence”ZDNet. Retrieved 1 March 2021.
  76. ^ Heath, Nick (11 December 2020). “What is AI? Everything you need to know about Artificial Intelligence”ZDNet. Retrieved 1 March 2021.
  77. ^ Kaplan, Andreas; Haenlein, Michael (1 January 2019). “Siri, Siri, in my hand: Who’s the fairest in the land? On the interpretations, illustrations, and implications of artificial intelligence”. Business Horizons62 (1): 15–25. doi:10.1016/j.bushor.2018.08.004.
  78. ^ Domingos 2015, Chapter 5.
  79. ^ Domingos 2015, Chapter 7.
  80. ^ Lindenbaum, M., Markovitch, S., & Rusakov, D. (2004). Selective sampling for nearest neighbor classifiers. Machine learning, 54(2), 125–152.
  81. ^ Domingos 2015, Chapter 1.
  82. a b Intractability and efficiency and the combinatorial explosion: * Russell & Norvig 2003, pp. 9, 21–22
  83. ^ Domingos 2015, Chapter 2, Chapter 3.
  84. ^ Hart, P. E.; Nilsson, N. J.; Raphael, B. (1972). “Correction to “A Formal Basis for the Heuristic Determination of Minimum Cost Paths””. SIGART Newsletter (37): 28–29. doi:10.1145/1056777.1056779S2CID 6386648.
  85. ^ Domingos 2015, Chapter 2, Chapter 4, Chapter 6.
  86. ^ “Can neural network computers learn from experience, and if so, could they ever become what we would call ‘smart’?”Scientific American. 2018. Archived from the original on 25 March 2018. Retrieved 24 March 2018.
  87. ^ Domingos 2015, Chapter 6, Chapter 7.
  88. ^ Domingos 2015, p. 286.
  89. ^ “Single pixel change fools AI programs”BBC News. 3 November 2017. Archived from the original on 22 March 2018. Retrieved 12 March 2018.
  90. ^ “AI Has a Hallucination Problem That’s Proving Tough to Fix”WIRED. 2018. Archived from the original on 12 March 2018. Retrieved 12 March 2018.
  91. ^ “Cultivating Common Sense | DiscoverMagazine.com”Discover Magazine. 2017. Archived from the original on 25 March 2018. Retrieved 24 March 2018.
  92. ^ Davis, Ernest; Marcus, Gary (24 August 2015). “Commonsense reasoning and commonsense knowledge in artificial intelligence”Communications of the ACM58 (9): 92–103. doi:10.1145/2701413S2CID 13583137Archived from the original on 22 August 2020. Retrieved 6 April 2020.
  93. ^ Winograd, Terry (January 1972). “Understanding natural language”. Cognitive Psychology3 (1): 1–191. doi:10.1016/0010-0285(72)90002-3.
  94. ^ “Don’t worry: Autonomous cars aren’t coming tomorrow (or next year)”Autoweek. 2016. Archived from the original on 25 March 2018. Retrieved 24 March 2018.
  95. ^ Knight, Will (2017). “Boston may be famous for bad drivers, but it’s the testing ground for a smarter self-driving car”MIT Technology ReviewArchived from the original on 22 August 2020. Retrieved 27 March 2018.
  96. ^ Prakken, Henry (31 August 2017). “On the problem of making autonomous vehicles conform to traffic law”Artificial Intelligence and Law25 (3): 341–363. doi:10.1007/s10506-017-9210-0.
  97. a b Lieto, Antonio; Lebiere, Christian; Oltramari, Alessandro (May 2018). “The knowledge level in cognitive architectures: Current limitations and possible developments”. Cognitive Systems Research48: 39–55. doi:10.1016/j.cogsys.2017.05.001hdl:2318/1665207S2CID 206868967.
  98. ^ Problem solving, puzzle solving, game playing and deduction: * Russell & Norvig 2003, chpt. 3–9, * Poole, Mackworth & Goebel 1998, chpt. 2,3,7,9, * Luger & Stubblefield 2004, chpt. 3,4,6,8, * Nilsson 1998, chpt. 7–12
  99. ^ Uncertain reasoning: * Russell & Norvig 2003, pp. 452–644, * Poole, Mackworth & Goebel 1998, pp. 345–395, * Luger & Stubblefield 2004, pp. 333–381, * Nilsson 1998, chpt. 19
  100. ^ Psychological evidence of sub-symbolic reasoning: * Wason & Shapiro (1966)showed that people do poorly on completely abstract problems, but if the problem is restated to allow the use of intuitive social intelligence, performance dramatically improves. (See Wason selection task) * Kahneman, Slovic & Tversky (1982) have shown that people are terrible at elementary problems that involve uncertain reasoning. (See list of cognitive biases for several examples). * Lakoff & Núñez (2000) have controversially argued that even our skills at mathematics depend on knowledge and skills that come from “the body”, i.e. sensorimotor and perceptual skills. (See Where Mathematics Comes From)
  101. ^ Knowledge representation: * ACM 1998, I.2.4, * Russell & Norvig 2003, pp. 320–363, * Poole, Mackworth & Goebel 1998, pp. 23–46, 69–81, 169–196, 235–277, 281–298, 319–345, * Luger & Stubblefield 2004, pp. 227–243, * Nilsson 1998, chpt. 18
  102. ^ Knowledge engineering: * Russell & Norvig 2003, pp. 260–266, * Poole, Mackworth & Goebel 1998, pp. 199–233, * Nilsson 1998, chpt. ≈17.1–17.4
  103. ^ Representing categories and relations: Semantic networksdescription logicsinheritance (including frames and scripts): * Russell & Norvig 2003, pp. 349–354, * Poole, Mackworth & Goebel 1998, pp. 174–177, * Luger & Stubblefield 2004, pp. 248–258, * Nilsson 1998, chpt. 18.3
  104. ^ Representing events and time:Situation calculusevent calculusfluent calculus(including solving the frame problem): * Russell & Norvig 2003, pp. 328–341, * Poole, Mackworth & Goebel 1998, pp. 281–298, * Nilsson 1998, chpt. 18.2
  105. ^ Causal calculus: * Poole, Mackworth & Goebel 1998, pp. 335–337
  106. ^ Representing knowledge about knowledge: Belief calculus, modal logics: * Russell & Norvig 2003, pp. 341–344, * Poole, Mackworth & Goebel 1998, pp. 275–277
  107. ^ Sikos, Leslie F. (June 2017). Description Logics in Multimedia Reasoning. Cham: Springer. doi:10.1007/978-3-319-54066-5ISBN 978-3-319-54066-5S2CID 3180114Archived from the original on 29 August 2017.
  108. ^ Ontology: * Russell & Norvig 2003, pp. 320–328
  109. ^ Smoliar, Stephen W.; Zhang, HongJiang (1994). “Content based video indexing and retrieval”. IEEE Multimedia1 (2): 62–72. doi:10.1109/93.311653S2CID 32710913.
  110. ^ Neumann, Bernd; Möller, Ralf (January 2008). “On scene interpretation with description logics”. Image and Vision Computing26 (1): 82–101. doi:10.1016/j.imavis.2007.08.013.
  111. ^ Kuperman, G. J.; Reichley, R. M.; Bailey, T. C. (1 July 2006). “Using Commercial Knowledge Bases for Clinical Decision Support: Opportunities, Hurdles, and Recommendations”Journal of the American Medical Informatics Association13(4): 369–371. doi:10.1197/jamia.M2055PMC 1513681PMID 16622160.
  112. ^ MCGARRY, KEN (1 December 2005). “A survey of interestingness measures for knowledge discovery”. The Knowledge Engineering Review20 (1): 39–61. doi:10.1017/S0269888905000408S2CID 14987656.
  113. ^ Bertini, M; Del Bimbo, A; Torniai, C (2006). “Automatic annotation and semantic retrieval of video sequences using multimedia ontologies”. MM ’06 Proceedings of the 14th ACM international conference on Multimedia. 14th ACM international conference on Multimedia. Santa Barbara: ACM. pp. 679–682.
  114. ^ Qualification problem: * McCarthy & Hayes 1969 * Russell & Norvig 2003[page needed] While McCarthy was primarily concerned with issues in the logical representation of actions, Russell & Norvig 2003 apply the term to the more general issue of default reasoning in the vast network of assumptions underlying all our commonsense knowledge.
  115. ^ Default reasoning and default logicnon-monotonic logicscircumscriptionclosed world assumptionabduction (Poole et al. places abduction under “default reasoning”. Luger et al. places this under “uncertain reasoning”): * Russell & Norvig 2003, pp. 354–360, * Poole, Mackworth & Goebel 1998, pp. 248–256, 323–335, * Luger & Stubblefield 2004, pp. 335–363, * Nilsson 1998, ~18.3.3
  116. ^ Breadth of commonsense knowledge: * Russell & Norvig 2003, p. 21, * Crevier 1993, pp. 113–114, * Moravec 1988, p. 13, * Lenat & Guha 1989 (Introduction)
  117. ^ Dreyfus & Dreyfus 1986.
  118. ^ Gladwell 2005.
  119. a b Expert knowledge as embodied intuition: * Dreyfus & Dreyfus 1986 (Hubert Dreyfus is a philosopher and critic of AI who was among the first to argue that most useful human knowledge was encoded sub-symbolically. See Dreyfus’ critique of AI) * Gladwell 2005 (Gladwell’s Blink is a popular introduction to sub-symbolic reasoning and knowledge.) * Hawkins & Blakeslee 2005 (Hawkins argues that sub-symbolic knowledge should be the primary focus of AI research.)
  120. ^ Planning: * ACM 1998, ~I.2.8, * Russell & Norvig 2003, pp. 375–459, * Poole, Mackworth & Goebel 1998, pp. 281–316, * Luger & Stubblefield 2004, pp. 314–329, * Nilsson 1998, chpt. 10.1–2, 22
  121. ^ Information value theory: * Russell & Norvig 2003, pp. 600–604
  122. ^ Classical planning: * Russell & Norvig 2003, pp. 375–430, * Poole, Mackworth & Goebel 1998, pp. 281–315, * Luger & Stubblefield 2004, pp. 314–329, * Nilsson 1998, chpt. 10.1–2, 22
  123. ^ Planning and acting in non-deterministic domains: conditional planning, execution monitoring, replanning and continuous planning: * Russell & Norvig 2003, pp. 430–449
  124. ^ Multi-agent planning and emergent behavior: * Russell & Norvig 2003, pp. 449–455
  125. ^ Turing 1950.
  126. ^ Solomonoff 1956.
  127. a b Learning: * ACM 1998, I.2.6, * Russell & Norvig 2003, pp. 649–788, * Poole, Mackworth & Goebel 1998, pp. 397–438, * Luger & Stubblefield 2004, pp. 385–542, * Nilsson 1998, chpt. 3.3, 10.3, 17.5, 20
  128. ^ Jordan, M. I.; Mitchell, T. M. (16 July 2015). “Machine learning: Trends, perspectives, and prospects”. Science349 (6245): 255–260. Bibcode:2015Sci…349..255Jdoi:10.1126/science.aaa8415PMID 26185243S2CID 677218.
  129. ^ Reinforcement learning: * Russell & Norvig 2003, pp. 763–788 * Luger & Stubblefield 2004, pp. 442–449
  130. ^ Natural language processing: * ACM 1998, I.2.7 * Russell & Norvig 2003, pp. 790–831 * Poole, Mackworth & Goebel 1998, pp. 91–104 * Luger & Stubblefield 2004, pp. 591–632
  131. ^ “Versatile question answering systems: seeing in synthesis” Archived 1 February 2016 at the Wayback Machine, Mittal et al., IJIIDS, 5(2), 119–142, 2011
  132. ^ Applications of natural language processing, including information retrieval (i.e. text mining) and machine translation: * Russell & Norvig 2003, pp. 840–857, * Luger & Stubblefield 2004, pp. 623–630
  133. ^ Cambria, Erik; White, Bebo (May 2014). “Jumping NLP Curves: A Review of Natural Language Processing Research [Review Article]”. IEEE Computational Intelligence Magazine9 (2): 48–57. doi:10.1109/MCI.2014.2307227S2CID 206451986.
  134. ^ Vincent, James (7 November 2019). “OpenAI has published the text-generating AI it said was too dangerous to share”The VergeArchived from the original on 11 June 2020. Retrieved 11 June 2020.
  135. ^ Machine perception: * Russell & Norvig 2003, pp. 537–581, 863–898 * Nilsson 1998, ~chpt. 6
  136. ^ Speech recognition: * ACM 1998, ~I.2.7 * Russell & Norvig 2003, pp. 568–578
  137. ^ Object recognition: * Russell & Norvig 2003, pp. 885–892
  138. ^ Computer vision: * ACM 1998, I.2.10 * Russell & Norvig 2003, pp. 863–898 * Nilsson 1998, chpt. 6
  139. ^ Robotics: * ACM 1998, I.2.9, * Russell & Norvig 2003, pp. 901–942, * Poole, Mackworth & Goebel 1998, pp. 443–460
  140. ^ Moving and configuration space: * Russell & Norvig 2003, pp. 916–932
  141. ^ Tecuci 2012.
  142. ^ Robotic mapping (localization, etc): * Russell & Norvig 2003, pp. 908–915
  143. ^ Cadena, Cesar; Carlone, Luca; Carrillo, Henry; Latif, Yasir; Scaramuzza, Davide; Neira, Jose; Reid, Ian; Leonard, John J. (December 2016). “Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age”. IEEE Transactions on Robotics32 (6): 1309–1332. arXiv:1606.05830Bibcode:2016arXiv160605830Cdoi:10.1109/TRO.2016.2624754S2CID 2596787.
  144. ^ Moravec 1988, p. 15.
  145. ^ Chan, Szu Ping (15 November 2015). “This is what will happen when robots take over the world”Archived from the original on 24 April 2018. Retrieved 23 April2018.
  146. ^ “IKEA furniture and the limits of AI”The Economist. 2018. Archived from the original on 24 April 2018. Retrieved 24 April 2018.
  147. ^ “Kismet”. MIT Artificial Intelligence Laboratory, Humanoid Robotics Group. Archived from the original on 17 October 2014. Retrieved 25 October 2014.
  148. ^ Thompson, Derek (2018). “What Jobs Will the Robots Take?”The AtlanticArchived from the original on 24 April 2018. Retrieved 24 April 2018.
  149. ^ Scassellati, Brian (2002). “Theory of mind for a humanoid robot”. Autonomous Robots12 (1): 13–24. doi:10.1023/A:1013298507114S2CID 1979315.
  150. ^ Cao, Yongcan; Yu, Wenwu; Ren, Wei; Chen, Guanrong (February 2013). “An Overview of Recent Progress in the Study of Distributed Multi-Agent Coordination”. IEEE Transactions on Industrial Informatics9 (1): 427–438. arXiv:1207.3231doi:10.1109/TII.2012.2219061S2CID 9588126.
  151. ^ Thro 1993.
  152. ^ Edelson 1991.
  153. ^ Tao & Tan 2005.
  154. ^ Poria, Soujanya; Cambria, Erik; Bajpai, Rajiv; Hussain, Amir (September 2017). “A review of affective computing: From unimodal analysis to multimodal fusion”. Information Fusion37: 98–125. doi:10.1016/j.inffus.2017.02.003hdl:1893/25490.
  155. ^ Emotion and affective computing: * Minsky 2006
  156. ^ Waddell, Kaveh (2018). “Chatbots Have Entered the Uncanny Valley”The AtlanticArchived from the original on 24 April 2018. Retrieved 24 April 2018.
  157. ^ Pennachin, C.; Goertzel, B. (2007). “Contemporary Approaches to Artificial General Intelligence”. Artificial General Intelligence. Cognitive Technologies. Berlin, Heidelberg: Springer. doi:10.1007/978-3-540-68677-4_1ISBN 978-3-540-23733-4.
  158. a b c Roberts, Jacob (2016). “Thinking Machines: The Search for Artificial Intelligence”Distillations. Vol. 2 no. 2. pp. 14–23. Archived from the original on 19 August 2018. Retrieved 20 March 2018.
  159. ^ “The superhero of artificial intelligence: can this genius keep it in check?”the Guardian. 16 February 2016. Archived from the original on 23 April 2018. Retrieved 26 April 2018.
  160. ^ Mnih, Volodymyr; Kavukcuoglu, Koray; Silver, David; Rusu, Andrei A.; Veness, Joel; Bellemare, Marc G.; Graves, Alex; Riedmiller, Martin; Fidjeland, Andreas K.; Ostrovski, Georg; Petersen, Stig; Beattie, Charles; Sadik, Amir; Antonoglou, Ioannis; King, Helen; Kumaran, Dharshan; Wierstra, Daan; Legg, Shane; Hassabis, Demis (26 February 2015). “Human-level control through deep reinforcement learning”. Nature518 (7540): 529–533. Bibcode:2015Natur.518..529Mdoi:10.1038/nature14236PMID 25719670S2CID 205242740.
  161. ^ Sample, Ian (14 March 2017). “Google’s DeepMind makes AI program that can learn like a human”the GuardianArchived from the original on 26 April 2018. Retrieved 26 April 2018.
  162. ^ “From not working to neural networking”The Economist. 2016. Archived from the original on 31 December 2016. Retrieved 26 April 2018.
  163. ^ Russell & Norvig 2009, Chapter 27. AI: The Present and Future.
  164. ^ Domingos 2015, Chapter 9. The Pieces of the Puzzle Fall into Place.
  165. a b Artificial brain arguments: AI requires a simulation of the operation of the human brain * Russell & Norvig 2003, p. 957 * Crevier 1993, pp. 271 & 279 A few of the people who make some form of the argument: * Moravec 1988 * Kurzweil 2005, p. 262 * Hawkins & Blakeslee 2005 The most extreme form of this argument (the brain replacement scenario) was put forward by Clark Glymour in the mid-1970s and was touched on by Zenon Pylyshyn and John Searle in 1980.
  166. ^ Goertzel, Ben; Lian, Ruiting; Arel, Itamar; de Garis, Hugo; Chen, Shuo (December 2010). “A world survey of artificial brain projects, Part II: Biologically inspired cognitive architectures”. Neurocomputing74 (1–3): 30–49. doi:10.1016/j.neucom.2010.08.012.
  167. ^ Nilsson 1983, p. 10.
  168. ^ AI’s immediate precursors: * McCorduck 2004, pp. 51–107 * Crevier 1993, pp. 27–32 * Russell & Norvig 2003, pp. 15, 940 * Moravec 1988, p. 3
  169. ^ Haugeland 1985, pp. 112–117
  170. ^ Cognitive simulation, Newell and Simon, AI at CMU (then called Carnegie Tech): * McCorduck 2004, pp. 139–179, 245–250, 322–323 (EPAM) * Crevier 1993, pp. 145–149
  171. ^ Soar (history): * McCorduck 2004, pp. 450–451 * Crevier 1993, pp. 258–263
  172. ^ McCarthy and AI research at SAIL and SRI International: * McCorduck 2004, pp. 251–259 * Crevier 1993
  173. ^ AI research at Edinburgh and in France, birth of Prolog: * Crevier 1993, pp. 193–196 * Howe 1994
  174. ^ AI at MIT under Marvin Minsky in the 1960s : * McCorduck 2004, pp. 259–305 * Crevier 1993, pp. 83–102, 163–176 * Russell & Norvig 2003, p. 19
  175. ^ Cyc: * McCorduck 2004, p. 489, who calls it “a determinedly scruffy enterprise” * Crevier 1993, pp. 239–243 * Russell & Norvig 2003, p. 363−365 * Lenat & Guha 1989
  176. ^ Knowledge revolution: * McCorduck 2004, pp. 266–276, 298–300, 314, 421 * Russell & Norvig 2003, pp. 22–23
  177. ^ Frederick, Hayes-Roth; William, Murray; Leonard, Adelman. “Expert systems”. AccessSciencedoi:10.1036/1097-8542.248550.
  178. ^ Embodied approaches to AI: * McCorduck 2004, pp. 454–462 * Brooks 1990 * Moravec 1988
  179. ^ Weng et al. 2001.
  180. ^ Lungarella et al. 2003.
  181. ^ Asada et al. 2009.
  182. ^ Oudeyer 2010.
  183. ^ Revival of connectionism: * Crevier 1993, pp. 214–215 * Russell & Norvig 2003, p. 25
  184. ^ Computational intelligence * IEEE Computational Intelligence SocietyArchived 9 May 2008 at the Wayback Machine
  185. ^ Hutson, Matthew (16 February 2018). “Artificial intelligence faces reproducibility crisis”Science. pp. 725–726. Bibcode:2018Sci…359..725Hdoi:10.1126/science.359.6377.725Archived from the original on 29 April 2018. Retrieved 28 April 2018.
  186. ^ Norvig 2012.
  187. ^ Langley 2011.
  188. ^ Katz 2012.
  189. ^ The intelligent agent paradigm: * Russell & Norvig 2003, pp. 27, 32–58, 968–972 * Poole, Mackworth & Goebel 1998, pp. 7–21 * Luger & Stubblefield 2004, pp. 235–240 * Hutter 2005, pp. 125–126 The definition used in this article, in terms of goals, actions, perception and environment, is due to Russell & Norvig (2003). Other definitions also include knowledge and learning as additional criteria.
  190. ^ Agent architectureshybrid intelligent systems: * Russell & Norvig (2003, pp. 27, 932, 970–972) * Nilsson (1998, chpt. 25)
  191. ^ Hierarchical control system: * Albus 2002
  192. ^ Lieto, Antonio; Bhatt, Mehul; Oltramari, Alessandro; Vernon, David (May 2018). “The role of cognitive architectures in general artificial intelligence”. Cognitive Systems Research48: 1–3. doi:10.1016/j.cogsys.2017.08.003hdl:2318/1665249S2CID 36189683.
  193. a b Russell & Norvig 2009, p. 1.
  194. a b White Paper: On Artificial Intelligence – A European approach to excellence and trust (PDF). Brussels: European Commission. 2020. p. 1. Archived (PDF)from the original on 20 February 2020. Retrieved 20 February 2020.
  195. ^ “AI set to exceed human brain power”CNN. 9 August 2006. Archived from the original on 19 February 2008.
  196. ^ Using AI to predict flight delays Archived 20 November 2018 at the Wayback Machine, Ishti.org.
  197. ^ N. Aletras; D. Tsarapatsanis; D. Preotiuc-Pietro; V. Lampos (2016). “Predicting judicial decisions of the European Court of Human Rights: a Natural Language Processing perspective”PeerJ Computer Science2: e93. doi:10.7717/peerj-cs.93.
  198. ^ “The Economist Explains: Why firms are piling into artificial intelligence”The Economist. 31 March 2016. Archived from the original on 8 May 2016. Retrieved 19 May 2016.
  199. ^ Lohr, Steve (28 February 2016). “The Promise of Artificial Intelligence Unfolds in Small Steps”The New York TimesArchived from the original on 29 February 2016. Retrieved 29 February 2016.
  200. ^ Frangoul, Anmar (14 June 2019). “A Californian business is using A.I. to change the way we think about energy storage”CNBCArchived from the original on 25 July 2020. Retrieved 5 November 2019.
  201. ^ Wakefield, Jane (15 June 2016). “Social media ‘outstrips TV’ as news source for young people”BBC NewsArchived from the original on 24 June 2016.
  202. ^ Smith, Mark (22 July 2016). “So you think you chose to read this article?”BBC NewsArchived from the original on 25 July 2016.
  203. ^ Brown, Eileen. “Half of Americans do not believe deepfake news could target them online”ZDNetArchived from the original on 6 November 2019. Retrieved 3 December 2019.
  204. ^ Zola, Andrew (12 April 2019). “Interview Prep: 40 Artificial Intelligence Questions”Springboard Blog.
  205. ^ The Turing test:
    Turing’s original publication: * Turing 1950 Historical influence and philosophical implications: * Haugeland 1985, pp. 6–9 * Crevier 1993, p. 24 * McCorduck 2004, pp. 70–71 * Russell & Norvig 2003, pp. 2–3 and 948
  206. ^ Dartmouth proposal: * McCarthy et al. 1955 (the original proposal) * Crevier 1993, p. 49 (historical significance)
  207. ^ The physical symbol systems hypothesis: * Newell & Simon 1976, p. 116 * McCorduck 2004, p. 153 * Russell & Norvig 2003, p. 18
  208. ^ Dreyfus 1992, p. 156.
  209. ^ Dreyfus’ critique of artificial intelligence: * Dreyfus 1972Dreyfus & Dreyfus 1986 * Crevier 1993, pp. 120–132 * McCorduck 2004, pp. 211–239 * Russell & Norvig 2003, pp. 950–952,
  210. ^ Gödel 1951: in this lecture, Kurt Gödel uses the incompleteness theorem to arrive at the following disjunction: (a) the human mind is not a consistent finite machine, or (b) there exist Diophantine equations for which it cannot decide whether solutions exist. Gödel finds (b) implausible, and thus seems to have believed the human mind was not equivalent to a finite machine, i.e., its power exceeded that of any finite machine. He recognized that this was only a conjecture, since one could never disprove (b). Yet he considered the disjunctive conclusion to be a “certain fact”.
  211. ^ The Mathematical Objection: * Russell & Norvig 2003, p. 949 * McCorduck 2004, pp. 448–449 Making the Mathematical Objection: * Lucas 1961 * Penrose 1989Refuting Mathematical Objection: * Turing 1950 under “(2) The Mathematical Objection” * Hofstadter 1979 Background: * Gödel 1931, Church 1936, Kleene 1935, Turing 1937
  212. ^ Graham Oppy (20 January 2015). “Gödel’s Incompleteness Theorems”Stanford Encyclopedia of PhilosophyArchived from the original on 22 April 2016. Retrieved 27 April 2016. These Gödelian anti-mechanist arguments are, however, problematic, and there is wide consensus that they fail.
  213. ^ Stuart J. RussellPeter Norvig (2010). “26.1.2: Philosophical Foundations/Weak AI: Can Machines Act Intelligently?/The mathematical objection”. Artificial Intelligence: A Modern Approach (3rd ed.). Upper Saddle River, NJ: Prentice HallISBN 978-0-13-604259-4even if we grant that computers have limitations on what they can prove, there is no evidence that humans are immune from those limitations.
  214. ^ Mark Colyvan. An introduction to the philosophy of mathematics. Cambridge University Press, 2012. From 2.2.2, ‘Philosophical significance of Gödel’s incompleteness results’: “The accepted wisdom (with which I concur) is that the Lucas-Penrose arguments fail.”
  215. ^ Iphofen, Ron; Kritikos, Mihalis (3 January 2019). “Regulating artificial intelligence and robotics: ethics by design in a digital society”. Contemporary Social Science: 1–15. doi:10.1080/21582041.2018.1563803ISSN 2158-2041.
  216. ^ “Ethical AI Learns Human Rights Framework”Voice of AmericaArchivedfrom the original on 11 November 2019. Retrieved 10 November 2019.
  217. ^ Crevier 1993, pp. 132–144.
  218. ^ Joseph Weizenbaum‘s critique of AI: * Weizenbaum 1976 * Crevier 1993, pp. 132–144 * McCorduck 2004, pp. 356–373 * Russell & Norvig 2003, p. 961 Weizenbaum (the AI researcher who developed the first chatterbot program, ELIZA) argued in 1976 that the misuse of artificial intelligence has the potential to devalue human life.
  219. ^ Wallach, Wendell (2010). Moral Machines. Oxford University Press.
  220. ^ Wallach 2010, pp. 37–54.
  221. ^ Wallach 2010, pp. 55–73.
  222. ^ Wallach 2010, “Introduction”.
  223. a b Michael Anderson and Susan Leigh Anderson (2011), Machine Ethics, Cambridge University Press.
  224. a b “Machine Ethics”aaai.org. Archived from the original on 29 November 2014.
  225. ^ Rubin, Charles (Spring 2003). “Artificial Intelligence and Human Nature”The New Atlantis1: 88–100. Archived from the original on 11 June 2012.
  226. ^ Brooks, Rodney (10 November 2014). “artificial intelligence is a tool, not a threat”. Archived from the original on 12 November 2014.
  227. ^ “Stephen Hawking, Elon Musk, and Bill Gates Warn About Artificial Intelligence”Observer. 19 August 2015. Archived from the original on 30 October 2015. Retrieved 30 October 2015.
  228. ^ Chalmers, David (1995). “Facing up to the problem of consciousness”Journal of Consciousness Studies2 (3): 200–219. Archived from the original on 8 March 2005. Retrieved 11 October 2018. See also this link Archived 8 April 2011 at the Wayback Machine
  229. ^ Horst, Steven, (2005) “The Computational Theory of Mind” Archived 11 September 2018 at the Wayback Machine in The Stanford Encyclopedia of Philosophy
  230. ^ Searle 1980, p. 1.
  231. ^ Searle’s Chinese room argument: * Searle 1980. Searle’s original presentation of the thought experiment. * Searle 1999. Discussion: * Russell & Norvig 2003, pp. 958–960 * McCorduck 2004, pp. 443–445 * Crevier 1993, pp. 269–271
  232. ^ Robot rights: * Russell & Norvig 2003, p. 964 Prematurity of: * Henderson 2007 In fiction: * McCorduck (2004, pp. 190–25) discusses Frankenstein and identifies the key ethical issues as scientific hubris and the suffering of the monster, i.e. robot rights.
  233. ^ “Robots could demand legal rights”BBC News. 21 December 2006. Archivedfrom the original on 15 October 2019. Retrieved 3 February 2011.
  234. ^ Evans, Woody (2015). “Posthuman Rights: Dimensions of Transhuman Worlds”Teknokultura12 (2). doi:10.5209/rev_TK.2015.v12.n2.49072.
  235. ^ maschafilm. “Content: Plug & Pray Film – Artificial Intelligence – Robots -“plugandpray-film.deArchived from the original on 12 February 2016.
  236. ^ Omohundro, Steve (2008). The Nature of Self-Improving Artificial Intelligence. presented and distributed at the 2007 Singularity Summit, San Francisco, CA.
  237. a b c Technological singularity: * Vinge 1993 * Kurzweil 2005 * Russell & Norvig 2003, p. 963
  238. ^ Transhumanism: * Moravec 1988 * Kurzweil 2005 * Russell & Norvig 2003, p. 963
  239. ^ AI as evolution: * Edward Fredkin is quoted in McCorduck (2004, p. 401). * Butler 1863 * Dyson 1998
  240. ^ “Robots and Artificial Intelligence”www.igmchicago.orgArchived from the original on 1 May 2019. Retrieved 3 July 2019.
  241. ^ “Sizing the prize: PwC’s Global AI Study—Exploiting the AI Revolution” (PDF). Retrieved 11 November 2020.
  242. ^ E McGaughey, ‘Will Robots Automate Your Job Away? Full Employment, Basic Income, and Economic Democracy’ (2018) SSRN, part 2(3) Archived 24 May 2018 at the Wayback Machine
  243. ^ “Automation and anxiety”The Economist. 9 May 2015. Archived from the original on 12 January 2018. Retrieved 13 January 2018.
  244. ^ Lohr, Steve (2017). “Robots Will Take Jobs, but Not as Fast as Some Fear, New Report Says”The New York TimesArchived from the original on 14 January 2018. Retrieved 13 January 2018.
  245. ^ Frey, Carl Benedikt; Osborne, Michael A (1 January 2017). “The future of employment: How susceptible are jobs to computerisation?”. Technological Forecasting and Social Change114: 254–280. CiteSeerX 10.1.1.395.416doi:10.1016/j.techfore.2016.08.019ISSN 0040-1625.
  246. ^ Arntz, Melanie, Terry Gregory, and Ulrich Zierahn. “The risk of automation for jobs in OECD countries: A comparative analysis.” OECD Social, Employment, and Migration Working Papers 189 (2016). p. 33.
  247. ^ Mahdawi, Arwa (26 June 2017). “What jobs will still be around in 20 years? Read this to prepare your future”The GuardianArchived from the original on 14 January 2018. Retrieved 13 January 2018.
  248. ^ Simon, Matt (1 April 2019). “Andrew Yang’s Presidential Bid Is So Very 21st Century”WiredArchived from the original on 24 June 2019. Retrieved 2 May2019 – via www.wired.com.
  249. ^ “Five experts share what scares them the most about AI”. 5 September 2018. Archived from the original on 8 December 2019. Retrieved 8 December 2019.
  250. ^ Russel, Stuart., Daniel Dewey, and Max Tegmark. Research Priorities for Robust and Beneficial Artificial Intelligence. AI Magazine 36:4 (2015). 8 December 2016.
  251. ^ “Commentary: Bad news. Artificial intelligence is biased”CNA. 12 January 2019. Archived from the original on 12 January 2019. Retrieved 19 June 2020.
  252. ^ Jeff Larson, Julia Angwin (23 May 2016). “How We Analyzed the COMPAS Recidivism Algorithm”ProPublicaArchived from the original on 29 April 2019. Retrieved 19 June 2020.
  253. ^ Rawlinson, Kevin (29 January 2015). “Microsoft’s Bill Gates insists AI is a threat”BBC NewsArchived from the original on 29 January 2015. Retrieved 30 January 2015.
  254. ^ Holley, Peter (28 January 2015). “Bill Gates on dangers of artificial intelligence: ‘I don’t understand why some people are not concerned'”The Washington PostISSN 0190-8286Archived from the original on 30 October 2015. Retrieved 30 October 2015.
  255. ^ Gibbs, Samuel (27 October 2014). “Elon Musk: artificial intelligence is our biggest existential threat”The GuardianArchived from the original on 30 October 2015. Retrieved 30 October 2015.
  256. ^ Churm, Philip Andrew (14 May 2019). “Yuval Noah Harari talks politics, technology and migration”euronews. Retrieved 15 November 2020.
  257. ^ Cellan-Jones, Rory (2 December 2014). “Stephen Hawking warns artificial intelligence could end mankind”BBC NewsArchived from the original on 30 October 2015. Retrieved 30 October 2015.
  258. ^ Bostrom, Nick (2015). “What happens when our computers get smarter than we are?”TED (conference)Archived from the original on 25 July 2020. Retrieved 30 January 2020.
  259. a b Russell, Stuart (8 October 2019). Human Compatible: Artificial Intelligence and the Problem of Control. United States: Viking. ISBN 978-0-525-55861-3OCLC 1083694322.
  260. ^ Post, Washington. “Tech titans like Elon Musk are spending $1 billion to save you from terminators”Archived from the original on 7 June 2016.
  261. ^ Müller, Vincent C.; Bostrom, Nick (2014). “Future Progress in Artificial Intelligence: A Poll Among Experts” (PDF). AI Matters1 (1): 9–11. doi:10.1145/2639475.2639478S2CID 8510016Archived (PDF) from the original on 15 January 2016.
  262. ^ “Oracle CEO Mark Hurd sees no reason to fear ERP AI”SearchERPArchived from the original on 6 May 2019. Retrieved 6 May 2019.
  263. ^ “Mark Zuckerberg responds to Elon Musk’s paranoia about AI: ‘AI is going to… help keep our communities safe.'”Business Insider. 25 May 2018. Archivedfrom the original on 6 May 2019. Retrieved 6 May 2019.
  264. ^ “The mysterious artificial intelligence company Elon Musk invested in is developing game-changing smart computers”Tech InsiderArchived from the original on 30 October 2015. Retrieved 30 October 2015.
  265. ^ Clark 2015a.
  266. ^ “Elon Musk Is Donating $10M Of His Own Money To Artificial Intelligence Research”Fast Company. 15 January 2015. Archived from the original on 30 October 2015. Retrieved 30 October 2015.
  267. ^ “Is artificial intelligence really an existential threat to humanity?”Bulletin of the Atomic Scientists. 9 August 2015. Archived from the original on 30 October 2015. Retrieved 30 October 2015.
  268. ^ “The case against killer robots, from a guy actually working on artificial intelligence”Fusion.netArchived from the original on 4 February 2016. Retrieved 31 January 2016.
  269. ^ “Will artificial intelligence destroy humanity? Here are 5 reasons not to worry”Vox. 22 August 2014. Archived from the original on 30 October 2015. Retrieved 30 October 2015.
  270. ^ Berryhill, Jamie; Heang, Kévin Kok; Clogher, Rob; McBride, Keegan (2019). Hello, World: Artificial Intelligence and its Use in the Public Sector (PDF). Paris: OECD Observatory of Public Sector Innovation. Archived (PDF) from the original on 20 December 2019. Retrieved 9 August 2020.
  271. ^ Barfield, Woodrow; Pagallo, Ugo (2018). Research handbook on the law of artificial intelligence. Cheltenham, UK. ISBN 978-1-78643-904-8OCLC 1039480085.
  272. ^ Law Library of Congress (U.S.). Global Legal Research Directorate, issuing body. Regulation of artificial intelligence in selected jurisdictionsLCCN 2019668143OCLC 1110727808.
  273. ^ Wirtz, Bernd W.; Weyerer, Jan C.; Geyer, Carolin (24 July 2018). “Artificial Intelligence and the Public Sector—Applications and Challenges”International Journal of Public Administration42 (7): 596–615. doi:10.1080/01900692.2018.1498103ISSN 0190-0692S2CID 158829602Archived from the original on 18 August 2020. Retrieved 22 August 2020.
  274. ^ Buiten, Miriam C (2019). “Towards Intelligent Regulation of Artificial Intelligence”European Journal of Risk Regulation10 (1): 41–59. doi:10.1017/err.2019.8ISSN 1867-299X.
  275. ^ Sotala, Kaj; Yampolskiy, Roman V (19 December 2014). “Responses to catastrophic AGI risk: a survey”Physica Scripta90 (1): 018001. doi:10.1088/0031-8949/90/1/018001ISSN 0031-8949.
  276. ^ Buttazzo, G. (July 2001). “Artificial consciousness: Utopia or real possibility?”. Computer34 (7): 24–30. doi:10.1109/2.933500.
  277. ^ Anderson, Susan Leigh. “Asimov’s “three laws of robotics” and machine metaethics.” AI & Society 22.4 (2008): 477–493.
  278. ^ McCauley, Lee (2007). “AI armageddon and the three laws of robotics”. Ethics and Information Technology9 (2): 153–164. CiteSeerX 10.1.1.85.8904doi:10.1007/s10676-007-9138-2S2CID 37272949.
  279. ^ Galvan, Jill (1 January 1997). “Entering the Posthuman Collective in Philip K. Dick’s “Do Androids Dream of Electric Sheep?””. Science Fiction Studies24 (3): 413–429. JSTOR 4240644.

AI textbooks

History of AI

Other sources

Further reading

  • DH Author, ‘Why Are There Still So Many Jobs? The History and Future of Workplace Automation’ (2015) 29(3) Journal of Economic Perspectives 3.
  • Boden, MargaretMind As MachineOxford University Press, 2006.
  • Cukier, Kenneth, “Ready for Robots? How to Think about the Future of AI”, Foreign Affairs, vol. 98, no. 4 (July/August 2019), pp. 192–98. George Dyson, historian of computing, writes (in what might be called “Dyson’s Law”) that “Any system simple enough to be understandable will not be complicated enough to behave intelligently, while any system complicated enough to behave intelligently will be too complicated to understand.” (p. 197.) Computer scientist Alex Pentland writes: “Current AI machine-learning algorithms are, at their core, dead simple stupid. They work, but they work by brute force.” (p. 198.)
  • Domingos, Pedro, “Our Digital Doubles: AI will serve our species, not control it”, Scientific American, vol. 319, no. 3 (September 2018), pp. 88–93.
  • Gopnik, Alison, “Making AI More Human: Artificial intelligence has staged a revival by starting to incorporate what we know about how children learn”, Scientific American, vol. 316, no. 6 (June 2017), pp. 60–65.
  • Johnston, John (2008) The Allure of Machinic Life: Cybernetics, Artificial Life, and the New AI, MIT Press.
  • Koch, Christof, “Proust among the Machines”, Scientific American, vol. 321, no. 6 (December 2019), pp. 46–49. Christof Koch doubts the possibility of “intelligent” machines attaining consciousness, because “[e]ven the most sophisticated brain simulations are unlikely to produce conscious feelings.” (p. 48.) According to Koch, “Whether machines can become sentient [is important] for ethical reasons. If computers experience life through their own senses, they cease to be purely a means to an end determined by their usefulness to… humans. Per GNW [the Global Neuronal Workspacetheory], they turn from mere objects into subjects… with a point of view…. Once computers’ cognitive abilities rival those of humanity, their impulse to push for legal and political rights will become irresistible – the right not to be deleted, not to have their memories wiped clean, not to suffer pain and degradation. The alternative, embodied by IIT [Integrated Information Theory], is that computers will remain only supersophisticated machinery, ghostlike empty shells, devoid of what we value most: the feeling of life itself.” (p. 49.)
  • Marcus, Gary, “Am I Human?: Researchers need new ways to distinguish artificial intelligence from the natural kind”, Scientific American, vol. 316, no. 3 (March 2017), pp. 58–63. A stumbling block to AI has been an incapacity for reliable disambiguation. An example is the “pronoun disambiguation problem”: a machine has no way of determining to whom or what a pronoun in a sentence refers. (p. 61.)
  • E McGaughey, ‘Will Robots Automate Your Job Away? Full Employment, Basic Income, and Economic Democracy’ (2018) SSRN, part 2(3) Archived 24 May 2018 at the Wayback Machine.
  • George Musser, “Artificial Imagination: How machines could learn creativity and common sense, among other human qualities”, Scientific American, vol. 320, no. 5 (May 2019), pp. 58–63.
  • Myers, Courtney Boyd ed. (2009). “The AI Report” Archived 29 July 2017 at the Wayback MachineForbes June 2009
  • Raphael, Bertram (1976). The Thinking Computer. W.H.Freeman and Company. ISBN 978-0-7167-0723-3Archived from the original on 26 July 2020. Retrieved 22 August 2020.
  • Scharre, Paul, “Killer Apps: The Real Dangers of an AI Arms Race”, Foreign Affairs, vol. 98, no. 3 (May/June 2019), pp. 135–44. “Today’s AI technologies are powerful but unreliable. Rules-based systems cannot deal with circumstances their programmers did not anticipate. Learning systems are limited by the data on which they were trained. AI failures have already led to tragedy. Advanced autopilot features in cars, although they perform well in some circumstances, have driven cars without warning into trucks, concrete barriers, and parked cars. In the wrong situation, AI systems go from supersmart to superdumb in an instant. When an enemy is trying to manipulate and hack an AI system, the risks are even greater.” (p. 140.)
  • Serenko, Alexander (2010). “The development of an AI journal ranking based on the revealed preference approach” (PDF). Journal of Informetrics4 (4): 447–459. doi:10.1016/j.joi.2010.04.001Archived (PDF) from the original on 4 October 2013. Retrieved 24 August 2013.
  • Serenko, Alexander; Michael Dohan (2011). “Comparing the expert survey and citation impact journal ranking methods: Example from the field of Artificial Intelligence”(PDF). Journal of Informetrics5 (4): 629–649. doi:10.1016/j.joi.2011.06.002Archived (PDF) from the original on 4 October 2013. Retrieved 12 September 2013.
  • Tom Simonite (29 December 2014). “2014 in Computing: Breakthroughs in Artificial Intelligence”MIT Technology Review.
  • Sun, R. & Bookman, L. (eds.), Computational Architectures: Integrating Neural and Symbolic Processes. Kluwer Academic Publishers, Needham, MA. 1994.
  • Taylor, Paul, “Insanely Complicated, Hopelessly Inadequate” (review of Brian Cantwell SmithThe Promise of Artificial Intelligence: Reckoning and Judgment, MIT, October 2019, ISBN 978 0 262 04304 5, 157 pp.; Gary Marcus and Ernest Davis, Rebooting AI: Building Artificial Intelligence We Can Trust, Ballantine, September 2019, ISBN 978 1 5247 4825 8, 304 pp.; Judea Pearl and Dana Mackenzie, The Book of Why: The New Science of Cause and Effect, Penguin, May 2019, ISBN 978 0 14 198241 0, 418 pp.), London Review of Books, vol. 43, no. 2 (21 January 2021), pp. 37–39. Paul Taylor writes (p. 39): “Perhaps there is a limit to what a computer can do without knowing that it is manipulating imperfect representations of an external reality.”
  • Tooze, Adam, “Democracy and Its Discontents”, The New York Review of Books, vol. LXVI, no. 10 (6 June 2019), pp. 52–53, 56–57. “Democracy has no clear answer for the mindless operation of bureaucratic and technological power. We may indeed be witnessing its extension in the form of artificial intelligence and robotics. Likewise, after decades of dire warning, the environmental problem remains fundamentally unaddressed…. Bureaucratic overreach and environmental catastrophe are precisely the kinds of slow-moving existential challenges that democracies deal with very badly…. Finally, there is the threat du jour: corporations and the technologies they promote.” (pp. 56–57.)

External links

Artificial intelligenceat Wikipedia’s sister projects

Categories

” (WP)

Sources:

Fair Use Sources:

Categories
Data Science - Big Data Software Engineering

Blockchain

For other uses, see Block chain (disambiguation).

” (WP)

blockchain is a growing list of records, called blocks, that are linked using cryptography.[1][2][3][4] Each block contains a cryptographic hash of the previous block,[4] a timestamp, and transaction data (generally represented as a Merkle tree). By design, a blockchain is resistant to modification of its data. This is because once recorded, the data in any given block cannot be altered retroactively without alteration of all subsequent blocks.

For use as a distributed ledger, a blockchain is typically managed by a peer-to-peer network collectively adhering to a protocol for inter-node communication and validating new blocks. Although blockchain records are not unalterable, blockchains may be considered secure by design and exemplify a distributed computing system with high Byzantine fault tolerance. The blockchain has been described as “an open, distributed ledger that can record transactions between two parties efficiently and in a verifiable and permanent way”.[5]

The blockchain was invented by a person (or group of people) using the name Satoshi Nakamoto in 2008 to serve as the public transaction ledger of the cryptocurrency bitcoin.[3] The identity of Satoshi Nakamoto remains unknown to date. The invention of the blockchain for bitcoin made it the first digital currency to solve the double-spending problem without the need of a trusted authority or central server. The bitcoin design has inspired other applications[3][2] and blockchains that are readable by the public and are widely used by cryptocurrencies. The blockchain is considered a type of payment rail.[6] Private blockchains have been proposed for business use but Computerworld called the marketing of such privatized blockchains without a proper security model “snake oil“.[7] However, others have argued that permissioned blockchains, if carefully designed, may be more decentralized and therefore more secure in practice than permissionless ones.[4][8]

File:Bitcoin Block Data.svg
Bitcoin blockchain structure

History

BitcoinEthereum and Litecoin transactions per day (January 2011 – January 2021)

Cryptographer David Chaum first proposed a blockchain-like protocol in his 1982 dissertation “Computer Systems Established, Maintained, and Trusted by Mutually Suspicious Groups.”[9] Further work on a cryptographically secured chain of blocks was described in 1991 by Stuart Haber and W. Scott Stornetta.[4][10] They wanted to implement a system where document timestamps could not be tampered with. In 1992, Haber, Stornetta, and Dave Bayer incorporated Merkle trees to the design, which improved its efficiency by allowing several document certificates to be collected into one block.[4][11]

The first blockchain was conceptualized by a person (or group of people) known as Satoshi Nakamoto in 2008. Nakamoto improved the design in an important way using a Hashcash-like method to timestamp blocks without requiring them to be signed by a trusted party and introducing a difficulty parameter to stabilize rate with which blocks are added to the chain.[4] The design was implemented the following year by Nakamoto as a core component of the cryptocurrency bitcoin, where it serves as the public ledger for all transactions on the network.[3]

In August 2014, the bitcoin blockchain file size, containing records of all transactions that have occurred on the network, reached 20 GB (gigabytes).[12] In January 2015, the size had grown to almost 30 GB, and from January 2016 to January 2017, the bitcoin blockchain grew from 50 GB to 100 GB in size. The ledger size had exceeded 200 GiB by early 2020.[13]

The words block and chain were used separately in Satoshi Nakamoto’s original paper, but were eventually popularized as a single word, blockchain, by 2016.

According to Accenture, an application of the diffusion of innovations theory suggests that blockchains attained a 13.5% adoption rate within financial services in 2016, therefore reaching the early adopters phase.[14] Industry trade groups joined to create the Global Blockchain Forum in 2016, an initiative of the Chamber of Digital Commerce.

In May 2018, Gartner found that only 1% of CIOs indicated any kind of blockchain adoption within their organisations, and only 8% of CIOs were in the short-term “planning or [looking at] active experimentation with blockchain”.[15] For the year 2019 Gartner reported 5% of CIOs believed blockchain technology was a ‘game-changer’ for their business.[16]

Structure

Blockchain formation. The main chain (black) consists of the longest series of blocks from the genesis block (green) to the current block. Orphan blocks (purple) exist outside of the main chain.

A blockchain is a decentralizeddistributed, and oftentimes public, digital ledger consisting of records called blocks that is used to record transactions across many computers so that any involved block cannot be altered retroactively, without the alteration of all subsequent blocks.[3][17] This allows the participants to verify and audit transactions independently and relatively inexpensively.[18] A blockchain database is managed autonomously using a peer-to-peer network and a distributed timestamping server. They are authenticated by mass collaboration powered by collective self-interests.[19] Such a design facilitates robust workflow where participants’ uncertainty regarding data security is marginal. The use of a blockchain removes the characteristic of infinite reproducibility from a digital asset. It confirms that each unit of value was transferred only once, solving the long-standing problem of double spending. A blockchain has been described as a value-exchange protocol.[20] A blockchain can maintain title rights because, when properly set up to detail the exchange agreement, it provides a record that compels offer and acceptance.

Logically, a blockchain can be seen as consisting of several layers:[21]

Blocks

Blocks hold batches of valid transactions that are hashed and encoded into a Merkle tree.[3] Each block includes the cryptographic hash of the prior block in the blockchain, linking the two. The linked blocks form a chain.[3] This iterative process confirms the integrity of the previous block, all the way back to the initial block, which is known as the genesis block.[22]

Sometimes separate blocks can be produced concurrently, creating a temporary fork. In addition to a secure hash-based history, any blockchain has a specified algorithm for scoring different versions of the history so that one with a higher score can be selected over others. Blocks not selected for inclusion in the chain are called orphan blocks.[22] Peers supporting the database have different versions of the history from time to time. They keep only the highest-scoring version of the database known to them. Whenever a peer receives a higher-scoring version (usually the old version with a single new block added) they extend or overwrite their own database and retransmit the improvement to their peers. There is never an absolute guarantee that any particular entry will remain in the best version of the history forever. Blockchains are typically built to add the score of new blocks onto old blocks and are given incentives to extend with new blocks rather than overwrite old blocks. Therefore, the probability of an entry becoming superseded decreases exponentially[23] as more blocks are built on top of it, eventually becoming very low.[3][24]:ch. 08[25] For example, bitcoin uses a proof-of-work system, where the chain with the most cumulative proof-of-work is considered the valid one by the network. There are a number of methods that can be used to demonstrate a sufficient level of computation. Within a blockchain the computation is carried out redundantly rather than in the traditional segregated and parallel manner.[26]

Block time

The block time is the average time it takes for the network to generate one extra block in the blockchain. Some blockchains create a new block as frequently as every five seconds.[27] By the time of block completion, the included data becomes verifiable. In cryptocurrency, this is practically when the transaction takes place, so a shorter block time means faster transactions. The block time for Ethereum is set to between 14 and 15 seconds, while for bitcoin it is on average 10 minutes.[28]

Hard forks

This section is an excerpt from Fork (blockchain) § Hard fork[edit]

hard fork is a rule change such that the software validating according to the old rules will see the blocks produced according to the new rules as invalid. In case of a hard fork, all nodes meant to work in accordance with the new rules need to upgrade their software. If one group of nodes continues to use the old software while the other nodes use the new software, a permanent split can occur.

For example, Ethereum has hard-forked to “make whole” the investors in The DAO, which had been hacked by exploiting a vulnerability in its code. In this case, the fork resulted in a split creating Ethereum and Ethereum Classic chains. In 2014 the Nxt community was asked to consider a hard fork that would have led to a rollback of the blockchain records to mitigate the effects of a theft of 50 million NXT from a major cryptocurrency exchange. The hard fork proposal was rejected, and some of the funds were recovered after negotiations and ransom payment. Alternatively, to prevent a permanent split, a majority of nodes using the new software may return to the old rules, as was the case of bitcoin split on 12 March 2013.[29]A more recent hard-fork example is of Bitcoin in 2017, which resulted in a split creating Bitcoin Cash.[30] The network split was mainly due a disagreement in how to increase the transactions per second to accommodate for demand.[31]

Decentralization

By storing data across its peer-to-peer network, the blockchain eliminates a number of risks that come with data being held centrally.[3] The decentralized blockchain may use ad hoc message passing and distributed networking.

Peer-to-peer blockchain networks lack centralized points of vulnerability that computer crackers can exploit; likewise, it has no central point of failure. Blockchain security methods include the use of public-key cryptography.[32]:5 A public key (a long, random-looking string of numbers) is an address on the blockchain. Value tokens sent across the network are recorded as belonging to that address. A private key is like a password that gives its owner access to their digital assets or the means to otherwise interact with the various capabilities that blockchains now support. Data stored on the blockchain is generally considered incorruptible.[3]

Every node in a decentralized system has a copy of the blockchain. Data quality is maintained by massive database replication[33] and computational trust. No centralized “official” copy exists and no user is “trusted” more than any other.[32] Transactions are broadcast to the network using software. Messages are delivered on a best-effort basis. Mining nodes validate transactions,[22] add them to the block they are building, and then broadcast the completed block to other nodes.[24]:ch. 08 Blockchains use various time-stamping schemes, such as proof-of-work, to serialize changes.[34] Alternative consensus methods include proof-of-stake.[22] Growth of a decentralized blockchain is accompanied by the risk of centralization because the computer resources required to process larger amounts of data become more expensive.[35]

Openness

Open blockchains are more user-friendly than some traditional ownership records, which, while open to the public, still require physical access to view. Because all early blockchains were permissionless, controversy has arisen over the blockchain definition. An issue in this ongoing debate is whether a private system with verifiers tasked and authorized (permissioned) by a central authority should be considered a blockchain.[36][37][38][39][40] Proponents of permissioned or private chains argue that the term “blockchain” may be applied to any data structure that batches data into time-stamped blocks. These blockchains serve as a distributed version of multiversion concurrency control (MVCC) in databases.[41] Just as MVCC prevents two transactions from concurrently modifying a single object in a database, blockchains prevent two transactions from spending the same single output in a blockchain.[42]:30–31 Opponents say that permissioned systems resemble traditional corporate databases, not supporting decentralized data verification, and that such systems are not hardened against operator tampering and revision.[36][38] Nikolai Hampton of Computerworld said that “many in-house blockchain solutions will be nothing more than cumbersome databases,” and “without a clear security model, proprietary blockchains should be eyed with suspicion.”[7][43]

Permissionless

An advantage to an open, permissionless, or public, blockchain network is that guarding against bad actors is not required and no access control is needed.[23] This means that applications can be added to the network without the approval or trust of others, using the blockchain as a transport layer.[23]

Bitcoin and other cryptocurrencies currently secure their blockchain by requiring new entries to include a proof of work. To prolong the blockchain, bitcoin uses Hashcash puzzles. While Hashcash was designed in 1997 by Adam Back, the original idea was first proposed by Cynthia Dwork and Moni Naor and Eli Ponyatovski in their 1992 paper “Pricing via Processing or Combatting Junk Mail”.

In 2016, venture capital investment for blockchain-related projects was weakening in the USA but increasing in China.[44] Bitcoin and many other cryptocurrencies use open (public) blockchains. As of April 2018, bitcoin has the highest market capitalization.

Permissioned (private) blockchain

See also: Distributed ledger

Permissioned blockchains use an access control layer to govern who has access to the network.[45] In contrast to public blockchain networks, validators on private blockchain networks are vetted by the network owner. They do not rely on anonymous nodes to validate transactions nor do they benefit from the network effect.[citation needed] Permissioned blockchains can also go by the name of ‘consortium’ blockchains.[citation needed] It has been argued that permissioned blockchains can guarantee a certain level of decentralization, if carefully designed, as opposed to permissionless blockchains, which are often centralized in practice.[8]

Disadvantages of private blockchain

Nikolai Hampton pointed out in Computerworld that “There is also no need for a ’51 percent’ attack on a private blockchain, as the private blockchain (most likely) already controls 100 percent of all block creation resources. If you could attack or damage the blockchain creation tools on a private corporate server, you could effectively control 100 percent of their network and alter transactions however you wished.”[7] This has a set of particularly profound adverse implications during a financial crisis or debt crisis like the financial crisis of 2007–08, where politically powerful actors may make decisions that favor some groups at the expense of others,[46][47] and “the bitcoin blockchain is protected by the massive group mining effort. It’s unlikely that any private blockchain will try to protect records using gigawatts of computing power — it’s time consuming and expensive.”[7] He also said, “Within a private blockchain there is also no ‘race’; there’s no incentive to use more power or discover blocks faster than competitors. This means that many in-house blockchain solutions will be nothing more than cumbersome databases.”[7]

Blockchain analysis

The analysis of public blockchains has become increasingly important with the popularity of bitcoinEthereumlitecoin and other cryptocurrencies.[48] A blockchain, if it is public, provides anyone who wants access to observe and analyse the chain data, given one has the know-how. The process of understanding and accessing the flow of crypto has been an issue for many cryptocurrencies, crypto-exchanges and banks.[49][50] The reason for this is accusations of blockchain enabled cryptocurrencies enabling illicit dark market trade of drugs, weapons, money laundering etc.[51] A common belief has been that cryptocurrency is private and untraceable, thus leading many actors to use it for illegal purposes. This is changing and now specialised tech-companies provide blockchain tracking services, making crypto exchanges, law-enforcement and banks more aware of what is happening with crypto funds and fiat crypto exchanges. The development, some argue, has led criminals to prioritise use of new cryptos such as Monero.[52][53][54] The question is about public accessibility of blockchain data and the personal privacy of the very same data. It is a key debate in cryptocurrency and ultimately in blockchain.[55]

Uses

Bitcoin’s transactions are recorded on a publicly viewable blockchain.

Blockchain technology can be integrated into multiple areas. The primary use of blockchains today is as a distributed ledger for cryptocurrencies, most notably bitcoin. There are a few operational products maturing from proof of concept by late 2016.[44] Businesses have been thus far reluctant to place blockchain at the core of the business structure.[56] Although businesses have been reluctant to fully implement blockchain, many have begun testing the technology and are conducting low-level implementation to gauge its effects on organizational efficiency.

In 2019, it was estimated that around $2.9 billion were invested in blockchain technology, which represents an 89% increase from the year prior. Additionally, the International Data Corp has estimated that corporate investment into blockchain technology will reach $12.4 billion by 2022.[57] Furthermore, According to PricewaterhouseCoopers (PwC), the second-largest professional services network in the world, blockchain technology has the potential to generate an annual business value of more than $3 trillion by 2030. PwC’s estimate is further augmented by a 2018 study that they have conducted, in which PwC surveyed 600 business executives and determined that 84% have at least some exposure to utilizing blockchain technology, which indicts a significant demand and interest in blockchain technology.[58]

Individual use of blockchain technology has also greatly increased since 2016. According to statistics in 2020, there were more than 40 million blockchain wallets in 2020 in comparison to around 10 million blockchain wallets in 2016.[59]

Cryptocurrencies

Main article: Cryptocurrency

Most cryptocurrencies use blockchain technology to record transactions. For example, the bitcoin network and Ethereum network are both based on blockchain. On 8 May 2018 Facebook confirmed that it would open a new blockchain group[60] which would be headed by David Marcus, who previously was in charge of Messenger. Facebook’s planned cryptocurrency platform, Libra (now known as Diem), was formally announced on June 18, 2019.[61][62]

The criminal enterprise Silk Road, which operated on Tor, utilized cryptocurrency for payments, some of which the US federal government has seized through research on the blockchain and forfeiture.[63]

Governments have mixed policies on the legality of their citizens or banks owning cryptocurrencies. China implements blockchain technology in several industries including a national digital currency which launched in 2020.[64][65] In order to strengthen their respective currencies, Western governments including the European Union and the United States have initiated similar projects.[66]

Smart contracts

Main article: Smart contract

Blockchain-based smart contracts are proposed contracts that can be partially or fully executed or enforced without human interaction.[67] One of the main objectives of a smart contract is automated escrow. A key feature of smart contracts is that they do not need a trusted third party (such as a trustee) to act as an intermediary between contracting entities -the blockchain network executes the contract on its own. This may reduce friction between entities when transferring value and could subsequently open the door to a higher level of transaction automation.[68] An IMF staff discussion reported that smart contracts based on blockchain technology might reduce moral hazards and optimize the use of contracts in general. But “no viable smart contract systems have yet emerged.” Due to the lack of widespread use their legal status is unclear.[69][70]

Financial services

According to Reason, many banks have expressed interest in implementing distributed ledgers for use in banking and are cooperating with companies creating private blockchains,[71][72][73] and according to a September 2016 IBM study, this is occurring faster than expected.[74]

Banks are interested in this technology because it has potential to speed up back office settlement systems.[75]

Banks such as UBS are opening new research labs dedicated to blockchain technology in order to explore how blockchain can be used in financial services to increase efficiency and reduce costs.[76][77]

Berenberg, a German bank, believes that blockchain is an “overhyped technology” that has had a large number of “proofs of concept”, but still has major challenges, and very few success stories.[78]

In December 2018, Bitwala launched Europe’s first regulated blockchain banking solution that enables users to manage both their bitcoin and euro deposits in one place with the safety and convenience of a German bank account. The bank account is hosted by the Berlin-based solarisBank.[79]

Mojaloop is designed to deliver financial support to people living in areas underserved by banks. It of use to migrants sending remittances[80]

The blockchain has also given rise to Initial coin offerings (ICOs) as well as a new category of digital asset called Security Token Offerings (STOs), also sometimes referred to as Digital Security Offerings (DSOs).[81] STO/DSOs may be conducted privately or on a public, regulated stock exchange and are used to tokenize traditional assets such as company shares as well as more innovative ones like intellectual property, real estate, art, or individual products. A number of companies are active in this space providing services for compliant tokenization, private STOs, and public STOs.

Video games

A blockchain game CryptoKitties, launched in November 2017.[82] The game made headlines in December 2017 when a cryptokitty character – an in-game virtual pet – was sold for more than US$100,000.[83] CryptoKitties illustrated scalability problems for games on Ethereum when it created significant congestion on the Ethereum network with about 30% of all Ethereum transactions being for the game.[84]

CryptoKitties also demonstrated how blockchains can be used to catalog game assets (digital assets).[85]

Energy trading

Blockchain is also being used in peer-to-peer energy trading.[86][87][88]

Supply chain

There are a number of efforts and industry organizations working to employ blockchains in supply chain management.

  • Mining — Blockchain technology allows wholesalers, retailers, and customers to track the origins of gemstones and other precious commodities. In 2016, The Wall Street Journal reported that the blockchain technology company, Everledger was partnering with IBM‘s blockchain-based tracking service to trace the origin of diamonds to insure that they were ethically mined.[89] DTC, the Diamond Trading Company has been involved in building a diamond trading supply chain product called Tracr.[90]
  • Food supply — Blockchain technology is being used to allow retailers and consumers to track the provenance of meat and other food products from their origins to stores and restaurants.[91] Walmart and IBM are running a trial to use a blockchain-backed system for supply chain monitoring for lettuce and spinach — all nodes of the blockchain are administered by Walmart and are located on the IBM cloud.[92] One cited benefit is that the system will enable rapid tracing of contaminated produce. Fogo de Chao, a Brazilian themed restaurant chain that features grilled meats, announced a partnership with HerdX, a blockchain-tech company focused on the food industry, that will enable suppliers, wholesalers, and diners to trace the beef served in Fogo de Chao restaurants back to the farm where it was raised.[93] Some analysts are less convinced that most consumers will be that interested in this capability.[90]
  • Shipping Walmart Canada uses a blockchain-based system developed by DLT Labs, a blockchain SaaS provider, that allows the retailer to track shipments and deliveries handled by dozens of third-party trucking companies. One reported benefit is that the blockchain-based system enables automated invoicing that reduces disputed billing, which in turn reduces delays in Walmart paying the freight transport companies.[94]
  • Blockchain software development — The Linux Foundation‘s blockchain initiative, Hyperledger Grid develops open components for blockchain supply chain solutions.[95][96] The goal of the project, said the foundation, was to “accelerate the development of blockchain-based solutions to cross-industry supply chain problems.”

Anti-counterfeiting

Blockchain could be used in detecting counterfeits by associating unique identifiers to products, documents and shipments, and storing records associated to transactions that cannot be forged or altered.[97][98] It is however argued that blockchain technology needs to be supplemented with technologies that provide a strong binding between physical objects and blockchain systems.[99] The EUIPO established an Anti-Counterfeiting Blockathon Forum, with the objective of “defining, piloting and implementing” an anti-counterfeiting infrastructure at the European level.[100][101] The Dutch Standardisation organisation NEN uses blockchain together with QR Codes to authenticate certificates.[102]

Healthcare

In response to the 2020 COVID-19 pandemicThe Wall Street Journal reported that Ernst & Young was working on a blockchain to help employers, governments, airlines and others keep track of people who have had antibody tests and could be immune to the virus. Hospitals and vendors also utilized a blockchain for needed medical equipment. Additionally, blockchain technology was being used in China to speed up the time it takes for health insurance payments to be paid to health-care providers and patients.[103]

Domain names

Blockchain domain names are another use of blockchain on the rise. Unlike regular domain names, blockchain domain names are entirely an asset of the domain owner and can only be controlled by the owner through a private key.[104] Blockchain domains pave the way to having sites that are more resistant to censorship and thus enable freedom of speech as there are no authorities or individuals that can intervene on controlling a domain except the private key holder.[105][106] They could be a better option to replace the traditional cryptocurrency wallet addresses as one can easily memorize the domain and use it for receiving payments.[107]

Organizations providing blockchain domain name services include Unstoppable Domains, Namecoin and Ethereum Name Services.[108]

Other uses

Blockchain technology can be used to create a permanent, public, transparent ledger system for compiling data on sales, tracking digital use and payments to content creators, such as wireless users[109] or musicians.[110] The Gartner 2019 CIO Survey reported 2% of higher education respondents had launched blockchain projects and another 18% were planning academic projects in the next 24 months.[111] In 2017, IBM partnered with ASCAP and PRS for Music to adopt blockchain technology in music distribution.[112] Imogen Heap‘s Mycelia service has also been proposed as blockchain-based alternative “that gives artists more control over how their songs and associated data circulate among fans and other musicians.”[113][114]

New distribution methods are available for the insurance industry such as peer-to-peer insuranceparametric insurance and microinsurance following the adoption of blockchain.[115][116] The sharing economy and IoT are also set to benefit from blockchains because they involve many collaborating peers.[117] Online voting is another application of the blockchain.[118][119] The use of blockchain in libraries is being studied with a grant from the U.S. Institute of Museum and Library Services.[120]

Other designs include:

  • Hyperledger is a cross-industry collaborative effort from the Linux Foundation to support blockchain-based distributed ledgers, with projects under this initiative including Hyperledger Burrow (by Monax) and Hyperledger Fabric (spearheaded by IBM).[121]
  • Quorum – a permissionable private blockchain by JPMorgan Chase with private storage, used for contract applications.[122]
  • Tezos, decentralized voting.[42]:94
  • Proof of Existence is an online service that verifies the existence of computer files as of a specific time.[123]

Types

Currently, there are at least four types of blockchain networks — public blockchains, private blockchains, consortium blockchains and hybrid blockchains.

Public blockchains

A public blockchain has absolutely no access restrictions. Anyone with an Internet connection can send transactions to it as well as become a validator (i.e., participate in the execution of a consensus protocol).[124][self-published source?] Usually, such networks offer economic incentives for those who secure them and utilize some type of a Proof of Stake or Proof of Work algorithm.

Some of the largest, most known public blockchains are the bitcoin blockchain and the Ethereum blockchain.

Private blockchains

A private blockchain is permissioned.[45] One cannot join it unless invited by the network administrators. Participant and validator access is restricted. To distinguish between open blockchains and other peer-to-peer decentralized database applications that are not open ad-hoc compute clusters, the terminology Distributed Ledger (DLT) is normally used for private blockchains.

Hybrid blockchains

A hybrid blockchain has a combination of centralized and decentralized features.[125] The exact workings of the chain can vary based on which portions of centralization decentralization are used.

Sidechains

A sidechain is a designation for a blockchain ledger that runs in parallel to a primary blockchain.[126][127] Entries from the primary blockchain (where said entries typically represent digital assets) can be linked to and from the sidechain; this allows the sidechain to otherwise operate independently of the primary blockchain (e.g., by using an alternate means of record keeping, alternate consensus algorithm, etc.).[128]

Interoperability

With the increasing number of blockchain systems appearing, even only those that support cryptocurrencies, blockchain interoperability is becoming a topic of major importance. The objective is to support transferring assets from one blockchain system to another blockchain system. Wegner[129] stated that “interoperability is the ability of two or more software components to cooperate despite differences in language, interface, and execution platform”. The objective of blockchain interoperability is therefore to support such cooperation among blockchain systems, despite those kinds of differences.

There are already several blockchain interoperability solutions available.[130] They can be classified in three categories: cryptocurrency interoperability approaches, blockchain engines, and blockchain connectors.

The IETF has a recent Blockchain-interop working group that already produced the draft of a blockchain interoperability architecture.[131]

Academic research

Blockchain panel discussion at the first IEEE Computer Society TechIgnite conference

In October 2014, the MIT Bitcoin Club, with funding from MIT alumni, provided undergraduate students at the Massachusetts Institute of Technology access to $100 of bitcoin. The adoption rates, as studied by Catalini and Tucker (2016), revealed that when people who typically adopt technologies early are given delayed access, they tend to reject the technology.[132]

Adoption decision

Motivations for adopting blockchain technology have been investigated by researchers. Janssen et al. provided a framework for analysis.[133] Koens & Poll pointed out that adoption could be heavily driven by non-technical factors.[134] Based on behavioral models, Li[135] discussed the differences between adoption at individual level and at organization level.

Collaboration

Scholars in business and management have started studying the role of blockchains to support collaboration.[136][137] It has been argued that blockchains can foster both cooperation (i.e., prevention of opportunistic behavior) and coordination (i.e., communication and information sharing). Thanks to reliability, transparency, traceability of records, and information immutability, blockchains facilitate collaboration in a way that differs both from the traditional use of contracts and from relational norms.[138] Contrary to contracts, blockchains do not directly rely on the legal system to enforce agreements.[139] In addition, contrary to the use of relational norms, blockchains do not require trust or direct connections between collaborators.

Blockchain and internal audit

External video
video icon Blockchain Basics & CryptographyGary GenslerMassachusetts Institute of Technology, 0:30[140]

The need for internal audit to provide effective oversight of organizational efficiency will require a change in the way that information is accessed in new formats.[141] Blockchain adoption requires a framework to identify the risk of exposure associated with transactions using blockchain. The Institute of Internal Auditors has identified the need for internal auditors to address this transformational technology. New methods are required to develop audit plans that identify threats and risks. The Internal Audit Foundation study, Blockchain and Internal Audit, assesses these factors.[142] The American Institute of Certified Public Accountants has outlined new roles for auditors as a result of blockchain.[143]

Energy use of proof-of-work blockchains

The Bank for International Settlements has criticized the public proof-of-work blockchains for high energy consumption.[144][145][146] Nicholas Weaver, of the International Computer Science Institute at the University of California, Berkeley examines blockchain’s online security, and the energy efficiency of proof-of-work public blockchains, and in both cases finds it grossly inadequate.[147][148] The 31—45 TWh of electricity used for bitcoin in 2018 produced 17—22.9 MtCO2.[149][150]

Journals

Main article: Ledger (journal)

In September 2015, the first peer-reviewed academic journal dedicated to cryptocurrency and blockchain technology research, Ledger, was announced. The inaugural issue was published in December 2016.[151] The journal covers aspects of mathematicscomputer scienceengineeringlaweconomics and philosophy that relate to cryptocurrencies such as bitcoin.[152][153]

The journal encourages authors to digitally sign a file hash of submitted papers, which are then timestamped into the bitcoin blockchain. Authors are also asked to include a personal bitcoin address in the first page of their papers for non-repudiation purposes.[154]

See also

References

  1. ^ Morris, David Z. (15 May 2016). “Leaderless, Blockchain-Based Venture Capital Fund Raises $100 Million, And Counting”FortuneArchived from the original on 21 May 2016. Retrieved 23 May 2016.
  2. a b Popper, Nathan (21 May 2016). “A Venture Fund With Plenty of Virtual Capital, but No Capitalist”The New York TimesArchived from the original on 22 May 2016. Retrieved 23 May 2016.
  3. a b c d e f g h i j “Blockchains: The great chain of being sure about things”The Economist. 31 October 2015. Archived from the original on 3 July 2016. Retrieved 18 June 2016. The technology behind bitcoin lets people who do not know or trust each other build a dependable ledger. This has implications far beyond the crypto currency.
  4. a b c d e f Narayanan, Arvind; Bonneau, Joseph; Felten, Edward; Miller, Andrew; Goldfeder, Steven (2016). Bitcoin and cryptocurrency technologies: a comprehensive introduction. Princeton: Princeton University Press. ISBN 978-0-691-17169-2.
  5. ^ Iansiti, Marco; Lakhani, Karim R. (January 2017). “The Truth About Blockchain”Harvard Business ReviewHarvard UniversityArchived from the original on 18 January 2017. Retrieved 17 January 2017. The technology at the heart of bitcoin and other virtual currencies, blockchain is an open, distributed ledger that can record transactions between two parties efficiently and in a verifiable and permanent way.
  6. ^ “Blockchain may finally disrupt payments from Micropayments to credit cards to SWIFT”dailyfintech.com. 10 February 2018. Archived from the original on 27 September 2018. Retrieved 18 November 2018.
  7. a b c d e Hampton, Nikolai (5 September 2016). “Understanding the blockchain hype: Why much of it is nothing more than snake oil and spin”ComputerworldArchived from the original on 6 September 2016. Retrieved 5 September 2016.
  8. a b Bakos, Yannis; Halaburda, Hanna; Mueller-Bloch, Christoph (February 2021). “When Permissioned Blockchains Deliver More Decentralization Than Permissionless”. Communications of the ACM64 (2): 20–22. doi:10.1145/3442371S2CID 231704491.
  9. ^ Sherman, Alan T.; Javani, Farid; Zhang, Haibin; Golaszewski, Enis (January 2019). “On the Origins and Variations of Blockchain Technologies”. IEEE Security Privacy17 (1): 72–77. arXiv:1810.06130doi:10.1109/MSEC.2019.2893730ISSN 1558-4046S2CID 53114747.
  10. ^ Haber, Stuart; Stornetta, W. Scott (January 1991). “How to time-stamp a digital document”. Journal of Cryptology3 (2): 99–111. CiteSeerX 10.1.1.46.8740doi:10.1007/bf00196791S2CID 14363020.
  11. ^ Bayer, Dave; Haber, Stuart; Stornetta, W. Scott (March 1992). Improving the Efficiency and Reliability of Digital Time-StampingSequences2. pp. 329–334. CiteSeerX 10.1.1.71.4891doi:10.1007/978-1-4613-9323-8_24ISBN 978-1-4613-9325-2.
  12. ^ Nian, Lam Pak; Chuen, David LEE Kuo (2015). “A Light Touch of Regulation for Virtual Currencies”. In Chuen, David LEE Kuo (ed.). Handbook of Digital Currency: Bitcoin, Innovation, Financial Instruments, and Big Data. Academic Press. p. 319. ISBN 978-0-12-802351-8.
  13. ^ “Blockchain Size”Archived from the original on 19 May 2020. Retrieved 25 February 2020.
  14. ^ “The future of blockchain in 8 charts”Raconteur. 27 June 2016. Archivedfrom the original on 2 December 2016. Retrieved 3 December 2016.
  15. ^ “Hype Killer – Only 1% of Companies Are Using Blockchain, Gartner Reports | Artificial Lawyer”Artificial Lawyer. 4 May 2018. Archived from the original on 22 May 2018. Retrieved 22 May 2018.
  16. ^ Kasey Panetta. (31 October 2018). “Digital Business: CIO Agenda 2019: Exploit Transformational Technologies.” Gartner website Retrieved 27 March 2021.
  17. ^ Armstrong, Stephen (7 November 2016). “Move over Bitcoin, the blockchain is only just getting started”WiredArchived from the original on 8 November 2016. Retrieved 9 November 2016.
  18. ^ Catalini, Christian; Gans, Joshua S. (23 November 2016). “Some Simple Economics of the Blockchain” (PDF). doi:10.2139/ssrn.2874598S2CID 46904163SSRN 2874598Archived (PDF) from the original on 6 March 2020. Retrieved 16 September 2019.
  19. ^ Tapscott, DonTapscott, Alex (8 May 2016). “Here’s Why Blockchains Will Change the World”FortuneArchived from the original on 13 November 2016. Retrieved 16 November 2016.
  20. ^ Bheemaiah, Kariappa (January 2015). “Block Chain 2.0: The Renaissance of Money”WiredArchived from the original on 14 November 2016. Retrieved 13 November 2016.
  21. ^ Chen, Huashan; Pendleton, Marcus; Njilla, Laurent; Xu, Shouhuai (12 June 2020). “A Survey on Ethereum Systems Security: Vulnerabilities, Attacks, and Defenses”ACM Computing Surveys53 (3): 3–4. arXiv:1908.04507doi:10.1145/3391195ISSN 0360-0300S2CID 199551841.
  22. a b c d Bhaskar, Nirupama Devi; Chuen, David LEE Kuo (2015). “Bitcoin Mining Technology”. Handbook of Digital Currency. pp. 45–65. doi:10.1016/B978-0-12-802117-0.00003-5ISBN 978-0-12-802117-0.
  23. a b c Antonopoulos, Andreas (20 February 2014). “Bitcoin security model: trust by computation”Radar. O’Reilly. Archived from the original on 31 October 2016. Retrieved 19 November 2016.
  24. a b Antonopoulos, Andreas M. (2014). Mastering Bitcoin. Unlocking Digital Cryptocurrencies. Sebastopol, CA: O’Reilly Media. ISBN 978-1449374037. Archived from the original on 1 December 2016. Retrieved 3 November 2015.
  25. ^ Nakamoto, Satoshi (October 2008). “Bitcoin: A Peer-to-Peer Electronic Cash System” (PDF). bitcoin.org. Archived (PDF) from the original on 20 March 2014. Retrieved 28 April 2014.
  26. ^ “Permissioned Blockchains”Explainer. Monax. Archived from the original on 20 November 2016. Retrieved 20 November 2016.
  27. ^ Strydom, Moses; Buckley, Sheryl (July 2019). AI and Big Data’s Potential for Disruptive Innovation. IGI Global. ISBN 978-1-5225-9687-5.
  28. ^ Kumar, Randhir; Tripathi, Rakesh (November 2019). “Implementation of Distributed File Storage and Access Framework using IPFS and Blockchain”. 2019 Fifth International Conference on Image Information Processing (ICIIP). IEEE: 246–251. doi:10.1109/iciip47207.2019.8985677ISBN 978-1-7281-0899-5S2CID 211119043.
  29. ^ Lee, Timothy (12 March 2013). “Major glitch in Bitcoin network sparks sell-off; price temporarily falls 23%”. Arstechnica. Archived from the original on 22 April 2013. Retrieved 25 February 2018.
  30. ^ Smith, Oli (21 January 2018). “Bitcoin price RIVAL: Cryptocurrency ‘faster than bitcoin’ will CHALLENGE market leaders”Express. Retrieved 6 April 2021.
  31. ^ “Bitcoin split in two, here’s what that means”CNN. 1 August 2017. Retrieved 7 April 2021.
  32. a b Brito, Jerry; Castillo, Andrea (2013). Bitcoin: A Primer for Policymakers(PDF) (Report). Fairfax, VA: Mercatus Center, George Mason University. Archived(PDF) from the original on 21 September 2013. Retrieved 22 October 2013.
  33. ^ Raval, Siraj (2016). Decentralized Applications: Harnessing Bitcoin’s Blockchain Technology. O’Reilly Media, Inc. pp. 12ISBN 978-1-4919-2452-5.
  34. ^ Kopfstein, Janus (12 December 2013). “The Mission to Decentralize the Internet”The New YorkerArchived from the original on 31 December 2014. Retrieved 30 December 2014. The network’s ‘nodes’ — users running the bitcoin software on their computers — collectively check the integrity of other nodes to ensure that no one spends the same coins twice. All transactions are published on a shared public ledger, called the ‘block chain.’
  35. ^ Gervais, Arthur; Karame, Ghassan O.; Capkun, Vedran; Capkun, Srdjan. “Is Bitcoin a Decentralized Currency?”InfoQ. InfoQ & IEEE computer society. Archived from the original on 10 October 2016. Retrieved 11 October 2016.
  36. a b Voorhees, Erik (30 October 2015). “It’s All About the Blockchain”Money and State. Archived from the original on 1 November 2015. Retrieved 2 November2015.
  37. ^ Reutzel, Bailey (13 July 2015). “A Very Public Conflict Over Private Blockchains”PaymentsSource. New York, NY: SourceMedia, Inc. Archivedfrom the original on 21 April 2016. Retrieved 18 June 2016.
  38. a b Casey, Michael J. (15 April 2015). “Moneybeat/BitBeat: Blockchains Without Coins Stir Tensions in Bitcoin Community”The Wall Street JournalArchivedfrom the original on 10 June 2016. Retrieved 18 June 2016.
  39. ^ “The ‘Blockchain Technology’ Bandwagon Has A Lesson Left To Learn”dinbits.com. 3 November 2015. Archived from the original on 29 June 2016. Retrieved 18 June 2016.
  40. ^ DeRose, Chris (26 June 2015). “Why the Bitcoin Blockchain Beats Out Competitors”American BankerArchived from the original on 30 March 2016. Retrieved 18 June 2016.
  41. ^ Greenspan, Gideon (19 July 2015). “Ending the bitcoin vs blockchain debate”multichain.comArchived from the original on 8 June 2016. Retrieved 18 June2016.
  42. a b Tapscott, DonTapscott, Alex (May 2016). The Blockchain Revolution: How the Technology Behind Bitcoin is Changing Money, Business, and the WorldISBN 978-0-670-06997-2.
  43. ^ Barry, Levine (11 June 2018). “A new report bursts the blockchain bubble”. MarTech. Archived from the original on 13 July 2018. Retrieved 13 July 2018.
  44. a b Ovenden, James. “Blockchain Top Trends In 2017”. The Innovation Enterprise. Archived from the original on 30 November 2016. Retrieved 4 December 2016.
  45. a b Bob Marvin (30 August 2017). “Blockchain: The Invisible Technology That’s Changing the World”PC MAG Australia. ZiffDavis, LLC. Archived from the original on 25 September 2017. Retrieved 25 September 2017.
  46. ^ Salsman, R.M. (19 September 2013). “The Financial Crisis Was A Failure Of Government, Not Free Markets”ForbesArchived from the original on 9 May 2018. Retrieved 8 May 2018.
  47. ^ O’Keeffe, M.; Terzi, A. (7 July 2015). “The political economy of financial crisis policy”BruegelArchived from the original on 19 May 2018. Retrieved 8 May2018.
  48. ^ Dr Garrick Hileman & Michel Rauchs (2017). “GLOBAL CRYPTOCURRENCY BENCHMARKING STUDY” (PDF). Cambridge Centre for Alternative Finance. University of Cambridge Judge Business School. Archived (PDF) from the original on 15 May 2019. Retrieved 15 May 2019 – via crowdfundinsider.
  49. ^ Raymaekers, Wim (March 2015). “Cryptocurrency Bitcoin: Disruption, challenges and opportunities”ingentaconnect.comArchived from the original on 15 May 2019. Retrieved 15 May 2019.
  50. ^ “Why Crypto Companies Still Can’t Open Checking Accounts”. 3 March 2019. Archived from the original on 4 June 2019. Retrieved 4 June 2019.
  51. ^ Christian Brenig, Rafael Accorsi & Günter Müller (Spring 2015). “Economic Analysis of Cryptocurrency Backed Money Laundering”Association for Information Systems AIS Electronic Library (AISeL)Archived from the original on 28 August 2019. Retrieved 15 May 2019.
  52. ^ Greenberg, Andy (25 January 2017). “Monero, the Drug Dealer’s Cryptocurrency of Choice, Is on Fire”WiredISSN 1059-1028Archived from the original on 10 December 2018. Retrieved 15 May 2019.
  53. ^ Orcutt, Mike. “It’s getting harder to hide money in Bitcoin”MIT Technology Review. Retrieved 15 May 2019.
  54. ^ “Explainer: ‘Privacy coin’ Monero offers near total anonymity”Reuters. 15 May 2019. Archived from the original on 15 May 2019. Retrieved 15 May 2019.
  55. ^ “An Untraceable Currency? Bitcoin Privacy Concerns – FinTech Weekly”FinTech Magazine Article. 7 April 2018. Archived from the original on 15 May 2019. Retrieved 15 May 2019.
  56. ^ Katie Martin (27 September 2016). “CLS dips into blockchain to net new currencies”Financial TimesArchived from the original on 9 November 2016. Retrieved 7 November 2016.
  57. ^ Michael Castillo (16 April 2019). “blockchain 50: Billion Dollar Babies”Financial Website. SourceMedia. Retrieved 1 February 2021.
  58. ^ Steve Davies (2018). “PwC’s Global Blockchain Survey”Financial Website. SourceMedia. Retrieved 1 February 2021.
  59. ^ Shanhong Liu (13 March 2020). “Blockchain – Statistics & Facts”Statistics Website. SourceMedia. Retrieved 17 February 2021.
  60. ^ Wagner, Kurt (8 May 2018). “Facebook is making its biggest executive shuffle in company history”RecodeArchived from the original on 22 July 2018. Retrieved 25 September 2018.
  61. ^ Isaac, Mike; Popper, Nathaniel (18 June 2019). “Facebook Plans Global Financial System Based on Cryptocurrency”The New York TimesArchived from the original on 19 May 2020. Retrieved 18 June 2019.
  62. ^ Constine, Josh (18 June 2019). “Facebook announces Libra cryptocurrency: All you need to know”TechCrunchArchived from the original on 19 June 2019. Retrieved 19 June 2019.
  63. ^ KPIX-TV. (5 November 2020). “Silk Road: Feds Seize $1 Billion In Bitcoins Linked To Infamous Silk Road Dark Web Case; ‘Where Did The Money Go’”. KPIX website Retrieved 28 March 2021.
  64. ^ Deloitte Insights. (2020). Deloitte’s 2020 Global blockchain technology: From promise to reality. Deloitte website Retrieved 31 March 2021.
  65. ^ Aditi Kumar and Eric Rosenbach. (20 May 2020). “Could China’s Digital Currency Unseat the Dollar?: American Economic and Geopolitical Power Is at Stake”. Foreign Affairs website Retrieved 31 March 2021.
  66. ^ Staff. (16 February 2021). “The Economist Explains: What is the fuss over central-bank digital currencies?” The Economist website Retrieved 1 April 2021.
  67. ^ Franco, Pedro (2014). Understanding Bitcoin: Cryptography, Engineering and Economics. John Wiley & Sons. p. 9. ISBN 978-1-119-01916-9Archived from the original on 14 February 2017. Retrieved 4 January 2017 – via Google Books.
  68. ^ Casey, Michael, 1967- (16 July 2018). The impact of blockchain technology on finance : a catalyst for change. London, UK. ISBN 978-1-912179-15-2OCLC 1059331326.
  69. ^ Governatori, Guido; Idelberger, Florian; Milosevic, Zoran; Riveret, Regis; Sartor, Giovanni; Xu, Xiwei (2018). “On legal contracts, imperative and declarative smart contracts, and blockchain systems”. Artificial Intelligence and Law26 (4): 33. doi:10.1007/s10506-018-9223-3S2CID 3663005.
  70. ^ Virtual Currencies and Beyond: Initial Considerations (PDF). IMF Discussion Note. International Monetary Fund. 2016. p. 23. ISBN 978-1-5135-5297-2Archived (PDF) from the original on 14 April 2018. Retrieved 19 April 2018.
  71. ^ Epstein, Jim (6 May 2016). “Is Blockchain Technology a Trojan Horse Behind Wall Street’s Walled Garden?”ReasonArchived from the original on 8 July 2016. Retrieved 29 June 2016. mainstream misgivings about working with a system that’s open for anyone to use. Many banks are partnering with companies building so-called private blockchains that mimic some aspects of Bitcoin’s architecture except they’re designed to be closed off and accessible only to chosen parties. … [but some believe] that open and permission-less blockchains will ultimately prevail even in the banking sector simply because they’re more efficient.
  72. ^ Redrup, Yolanda (29 June 2016). “ANZ backs private blockchain, but won’t go public”Australia Financial ReviewArchived from the original on 3 July 2016. Retrieved 7 July 2016. Blockchain networks can be either public or private. Public blockchains have many users and there are no controls over who can read, upload or delete the data and there are an unknown number of pseudonymous participants. In comparison, private blockchains also have multiple data sets, but there are controls in place over who can edit data and there are a known number of participants.
  73. ^ Shah, Rakesh (1 March 2018). “How Can The Banking Sector Leverage Blockchain Technology?”PostBox Communications. PostBox Communications Blog. Archived from the original on 17 March 2018. Banks preferably have a notable interest in utilizing Blockchain Technology because it is a great source to avoid fraudulent transactions. Blockchain is considered hassle free, because of the extra level of security it offers.
  74. ^ Kelly, Jemima (28 September 2016). “Banks adopting blockchain ‘dramatically faster’ than expected: IBM”ReutersArchived from the original on 28 September 2016. Retrieved 28 September 2016.
  75. ^ Arnold, Martin (23 September 2013). “IBM in blockchain project with China UnionPay”Financial TimesArchived from the original on 9 November 2016. Retrieved 7 November 2016.
  76. ^ “UBS leads team of banks working on blockchain settlement system”Reuters. 24 August 2016. Archived from the original on 19 May 2017. Retrieved 13 May2017.
  77. ^ “Cryptocurrency Blockchain”capgemini.comArchived from the original on 5 December 2016. Retrieved 13 May 2017.
  78. ^ Kelly, Jemima (31 October 2017). “Top banks and R3 build blockchain-based payments system”ReutersArchived from the original on 10 July 2018. Retrieved 9 July 2018.
  79. ^ Köln, Nils Wischmeyer (2018). “Bank mit Kette”Sueddeutsche Zeitung (in German). ISSN 0174-4917Archived from the original on 24 March 2020. Retrieved 29 April 2019.
  80. ^ “How Blockchain-based technology is disrupting migrants’ remittances: a preliminary assessment” (PDF).
  81. ^ “Archived copy” (PDF). Archived (PDF) from the original on 23 June 2019. Retrieved 26 September 2019.
  82. ^ “Internet firms try their luck at blockchain games”Asia Times. 22 February 2018. Retrieved 28 February 2018.
  83. ^ Evelyn Cheng (6 December 2017). “Meet CryptoKitties, the $100,000 digital beanie babies epitomizing the cryptocurrency mania”CNBCArchived from the original on 20 November 2018. Retrieved 28 February 2018.
  84. ^ Laignee Barron (13 February 2018). “CryptoKitties is Going Mobile. Can Ethereum Handle the Traffic?”FortuneArchived from the original on 28 October 2018. Retrieved 30 September 2018.
  85. ^ “CryptoKitties craze slows down transactions on Ethereum”. 12 May 2017. Archived from the original on 12 January 2018.
  86. ^ “Blockchain technology in the energy sector: A systematic review of challenges and opportunities”Archived from the original on 22 June 2020. Retrieved 7 June 2020.
  87. ^ “This Blockchain-Based Energy Platform Is Building A Peer-To-Peer Grid”Archived from the original on 7 June 2020. Retrieved 7 June 2020.
  88. ^ “Blockchain-based microgrid gives power to consumers in New York”Archived from the original on 22 March 2016. Retrieved 7 June 2020.
  89. ^ Nash, Kim S. (14 July 2016). “IBM Pushes Blockchain into the Supply Chain”The Wall Street JournalArchived from the original on 18 July 2016. Retrieved 24 July 2016.
  90. a b Gstettner, Stefan (30 July 2019). “How Blockchain Will Redefine Supply Chain Management”Knowledge@Wharton. The Wharton School of the University of Pennsylvania. Retrieved 28 August 2020.
  91. ^ Leong, Christine; Viskin, Tal; Stewart, Robyn (2018). “Blockchain Feasibility Study: Tracing the Supply Chain – How blockchain can enable traceability on the food industry” (PDF). Accenture.com. Accenture. Retrieved 18 August 2020.
  92. ^ Corkery, Michael; Popper, Nathaniel (24 September 2018). “From Farm to Blockchain: Walmart Tracks Its Lettuce”The New York TimesArchived from the original on 5 December 2018. Retrieved 5 December 2018.
  93. ^ Bandoim, Lana (30 April 2019). “Can Blockchain And Chip Technology Improve Beef Sourcing Transparency?”Forbes. Retrieved 19 August 2020.
  94. ^ Vitasek, Kate (31 January 2020). “Walmart Canada And DLT Labs Launch World’s Largest Industrial Blockchain Application”Forbes. Retrieved 28 August 2020.
  95. ^ Mearian, Lucas (23 January 2019). “Grid, a new project from the Linux Foundation, will offer developers tools to create supply chain-specific applications running atop distributed ledger technology”ComputerworldArchived from the original on 3 February 2019. Retrieved 8 March 2019.
  96. ^ Hyperledger (22 January 2019). “Announcing Hyperledger Grid, a new project to help build and deliver supply chain solutions!”Archived from the original on 4 February 2019. Retrieved 8 March 2019.
  97. ^ Ma, Jinhua; Lin, Shih-Ya; Chen, Xin; Sun, Hung-Min; Chen, Yeh-Cheng; Wang, Huaxiong (2020). “A Blockchain-Based Application System for Product Anti-Counterfeiting”IEEE Access8: 77642–77652. doi:10.1109/ACCESS.2020.2972026ISSN 2169-3536S2CID 214205788.
  98. ^ Alzahrani, Naif; Bulusu, Nirupama (15 June 2018). “Block-Supply Chain: A New Anti-Counterfeiting Supply Chain Using NFC and Blockchain”. Proceedings of the 1st Workshop on Cryptocurrencies and Blockchains for Distributed Systems. CryBlock’18. Munich, Germany: Association for Computing Machinery: 30–35. doi:10.1145/3211933.3211939ISBN 978-1-4503-5838-5S2CID 169188795.
  99. ^ Balagurusamy, V. S. K.; Cabral, C.; Coomaraswamy, S.; Delamarche, E.; Dillenberger, D. N.; Dittmann, G.; Friedman, D.; Gökçe, O.; Hinds, N.; Jelitto, J.; Kind, A. (1 March 2019). “Crypto anchors”IBM Journal of Research and Development63 (2/3): 4:1–4:12. doi:10.1147/JRD.2019.2900651ISSN 0018-8646.
  100. ^ Brett, Charles (18 April 2018). “EUIPO Blockathon Challenge 2018 -“Enterprise Times. Retrieved 1 September 2020.
  101. ^ “EUIPO Anti-Counterfeiting Blockathon Forum”.
  102. ^ “PT Industrieel Management”PT Industrieel Management. Retrieved 1 September 2020.
  103. ^ Castellanos, Sara. “A Cryptocurrency Technology Finds New Use Tackling Coronavirus”The Wall Street Journal. Retrieved 21 October 2020.
  104. ^ “Blockchain Domains: What Are They and How Are They Implemented? | Hacker Noon”hackernoon.com. Retrieved 16 April 2020.
  105. ^ “Unstoppable Domains and the End of Internet Censorship”HedgeTrade Blog. 22 April 2019. Archived from the original on 6 November 2019. Retrieved 16 April2020.
  106. ^ “Commentary: Blockchain Could Be the Savior of Free Speech”FortuneArchived from the original on 27 December 2019. Retrieved 16 April 2020.
  107. ^ “.Kred launches as dual DNS and ENS domain”Domain Name Wire | Domain Name News. 6 March 2020. Archived from the original on 8 March 2020. Retrieved 16 April 2020.
  108. ^ S, James; August 28, ers in Networking on; 2019; Pst, 7:26 Am. “Blockchain-based Unstoppable Domains is a rehash of a failed idea”TechRepublicArchived from the original on 19 November 2019. Retrieved 16 April 2020.
  109. ^ K. Kotobi, and S. G. Bilen, “Secure Blockchains for Dynamic Spectrum Access : A Decentralized Database in Moving Cognitive Radio Networks Enhances Security and User Access”, IEEE Vehicular Technology Magazine, 2018.
  110. ^ “Blockchain Could Be Music’s Next Disruptor”. 22 September 2016. Archivedfrom the original on 23 September 2016.
  111. ^ Susan Moore. (16 October 2019). “Digital Business: 4 Ways Blockchain Will Transform Higher Education”. Gartner website Retrieved 27 March 2021.
  112. ^ “ASCAP, PRS and SACEM Join Forces for Blockchain Copyright System”. Music Business Worldwide. 9 April 2017. Archived from the original on 10 April 2017.
  113. ^ Burchardi, K.; Harle, N. (20 January 2018). “The blockchain will disrupt the music business and beyond”Wired UKArchived from the original on 8 May 2018. Retrieved 8 May 2018.
  114. ^ Bartlett, Jamie (6 September 2015). “Imogen Heap: saviour of the music industry?”The GuardianArchived from the original on 22 April 2016. Retrieved 18 June 2016.
  115. ^ Wang, Kevin; Safavi, Ali (29 October 2016). “Blockchain is empowering the future of insurance”Tech Crunch. AOL Inc. Archived from the original on 7 November 2016. Retrieved 7 November 2016.
  116. ^ Gatteschi, Valentina; Lamberti, Fabrizio; Demartini, Claudio; Pranteda, Chiara; Santamaría, Víctor (20 February 2018). “Blockchain and Smart Contracts for Insurance: Is the Technology Mature Enough?”Future Internet10 (2): 20. doi:10.3390/fi10020020.
  117. ^ “Blockchain reaction: Tech companies plan for critical mass” (PDF). Ernst & Young. p. 5. Archived (PDF) from the original on 14 November 2016. Retrieved 13 November 2016.
  118. ^ “Online Voting Platform FAQ’s”Follow My VoteArchived from the original on 27 October 2016. Retrieved 7 December 2016.
  119. ^ Chandra, Prabhul. “Reimagining Democracy: What if votes were a crypto-currency?”democracywithoutborders.orgArchived from the original on 5 February 2018. Retrieved 5 February 2018.
  120. ^ Carrie Smith. Blockchain Reaction: How library professionals are approaching blockchain technology and its potential impact. Archived 12 September 2019 at the Wayback Machine American Libraries March 2019.
  121. ^ “IBM Blockchain based on Hyperledger Fabric from the Linux Foundation”IBM.com. 9 January 2018. Archived from the original on 7 December 2017. Retrieved 18 January 2018.
  122. ^ “Why J.P. Morgan Chase Is Building a Blockchain on Ethereum”FortuneArchived from the original on 2 February 2017. Retrieved 24 January 2017.
  123. ^ Melanie Swan (2015). Blockchain: Blueprint for a New EconomyO’Reilly Media. pp. 38–39. ISBN 9781491920473.
  124. ^ “How Companies Can Leverage Private Blockchains to Improve Efficiency and Streamline Business Processes”Perfectial.
  125. ^ [Distributed Ledger Technology: Hybrid Approach, Front-to-Back Designing and Changing Trade Processing Infrastructure, By Martin Walker, First published:, 24 OCT 2018 ISBN 978-1-78272-389-9]
  126. ^ Siraj Raval (18 July 2016). Decentralized Applications: Harnessing Bitcoin’s Blockchain Technology. “O’Reilly Media, Inc.”. pp. 22–. ISBN 978-1-4919-2452-5.
  127. ^ Niaz Chowdhury (16 August 2019). Inside Blockchain, Bitcoin, and Cryptocurrencies. CRC Press. pp. 22–. ISBN 978-1-00-050770-6.
  128. ^ U.S. Patent 10,438,290
  129. ^ Wegner, Peter (March 1996). “Interoperability”ACM Computing Surveys28: 285–287. doi:10.1145/234313.234424. Retrieved 24 October 2020.
  130. ^ Belchior, Rafael; Vasconcelos, André; Guerreiro, Sérgio; Correia, Miguel (May 2020). “A Survey on Blockchain Interoperability: Past, Present, and Future Trends”. arXiv:2005.14282 [cs.DC].
  131. ^ Hardjono, T.; Hargreaves, M.; Smith, N. (2 October 2020). An Interoperability Architecture for Blockchain Gateways (Technical report). IETF. draft-hardjono-blockchain-interop-arch-00.
  132. ^ Catalini, Christian; Tucker, Catherine E. (11 August 2016). “Seeding the S-Curve? The Role of Early Adopters in Diffusion”. doi:10.2139/ssrn.2822729S2CID 157317501SSRN 2822729.
  133. ^ Janssen, Marijn; Weerakkody, Vishanth; Ismagilova, Elvira; Sivarajah, Uthayasankar; Irani, Zahir (2020). “A framework for analysing blockchain technology adoption: Integrating institutional, market and technical factors”International Journal of Information Management. Elsevier. 50: 302–309. doi:10.1016/j.ijinfomgt.2019.08.012Archived from the original on 19 May 2020. Retrieved 16 April 2020.
  134. ^ Koens, Tommy; Poll, Erik (2019), “The Drivers Behind Blockchain Adoption: The Rationality of Irrational Choices”, Euro-Par 2018: Parallel Processing Workshops, Lecture Notes in Computer Science, 11339, pp. 535–546, doi:10.1007/978-3-030-10549-5_42ISBN 978-3-030-10548-8
  135. ^ Li, Jerry (2020), “Blockchain technology adoption: Examining the Fundamental Drivers”, Proceedings of the 2nd International Conference on Management Science and Industrial Engineering, ACM Publication, April 2020, pp. 253–260. https://dl.acm.org/doi/abs/10.1145/3396743.3396750 Archived 5 June 2020 at the Wayback Machine
  136. ^ Hsieh, Ying-Ying; Vergne, Jean-Philippe; Anderson, Philip; Lakhani, Karim; Reitzig, Markus (12 February 2019). “Correction to: Bitcoin and the rise of decentralized autonomous organizations”Journal of Organization Design8 (1): 3. doi:10.1186/s41469-019-0041-1ISSN 2245-408X.
  137. ^ Felin, Teppo; Lakhani, Karim (2018). “What Problems Will You Solve With Blockchain?”. MIT Sloan Management Review.
  138. ^ Lumineau, Fabrice; Wang, Wenqian; Schilke, Oliver (2020). “Blockchain Governance—A New Way of Organizing Collaborations?”. Organization Science32(2): 500–521. doi:10.1287/orsc.2020.1379.
  139. ^ Beck, Roman; Mueller-Bloch, Christoph; King, John Leslie (2018). “Governance in the Blockchain Economy: A Framework and Research Agenda”Journal of the Association for Information Systems: 1020–1034. doi:10.17705/1jais.00518.
  140. ^ Popper, Nathaniel (27 June 2018). “What is the Blockchain? Explaining the Tech Behind Cryptocurrencies (Published 2018)”The New York Times.
  141. ^ Hugh Rooney, Brian Aiken, & Megan Rooney. (2017). Q&A. Is Internal Audit Ready for Blockchain? Technology Innovation Management Review, (10), 41.
  142. ^ Richard C. Kloch, Jr Simon J. Little, Blockchain and Internal Audit Internal Audit Foundation, 2019 ISBN 978-1-63454-065-0
  143. ^ Alexander, A. (2019). The audit, transformed: New advancements in technology are reshaping this core service. Accounting Today, 33(1)
  144. ^ Hyun Song Shin (June 2018). “Chapter V. Cryptocurrencies: looking beyond the hype” (PDF). BIS 2018 Annual Economic Report. Bank for International Settlements. Archived (PDF) from the original on 18 June 2018. Retrieved 19 June 2018. Put in the simplest terms, the quest for decentralised trust has quickly become an environmental disaster.
  145. ^ Janda, Michael (18 June 2018). “Cryptocurrencies like bitcoin cannot replace money, says Bank for International Settlements”. ABC (Australia). Archived from the original on 18 June 2018. Retrieved 18 June 2018.
  146. ^ Hiltzik, Michael (18 June 2018). “Is this scathing report the death knell for bitcoin?”Los Angeles TimesArchived from the original on 18 June 2018. Retrieved 19 June 2018.
  147. ^ Illing, Sean (11 April 2018). “Why Bitcoin is bullshit, explained by an expert”VoxArchived from the original on 17 July 2018. Retrieved 17 July 2018.
  148. ^ Weaver, Nicholas. “Blockchains and Cryptocurrencies: Burn It With Fire”YouTube video. Berkeley School of Information. Archived from the original on 19 February 2019. Retrieved 17 July 2018.
  149. ^ Köhler, Susanne; Pizzol, Massimo (20 November 2019). “Life Cycle Assessment of Bitcoin Mining”Environmental Science & Technology53 (23): 13598–13606. Bibcode:2019EnST…5313598Kdoi:10.1021/acs.est.9b05687PMID 31746188.
  150. ^ Stoll, Christian; Klaaßen, Lena; Gallersdörfer, Ulrich (2019). “The Carbon Footprint of Bitcoin”Joule3 (7): 1647–1661. doi:10.1016/j.joule.2019.05.012.
  151. ^ Extance, Andy (30 September 2015). “The future of cryptocurrencies: Bitcoin and beyond”Nature526 (7571): 21–23. Bibcode:2015Natur.526…21Edoi:10.1038/526021aISSN 0028-0836OCLC 421716612PMID 26432223.
  152. ^ Ledger (eJournal / eMagazine, 2015). OCLC. OCLC 910895894.
  153. ^ Hertig, Alyssa (15 September 2015). “Introducing Ledger, the First Bitcoin-Only Academic Journal”MotherboardArchived from the original on 10 January 2017. Retrieved 10 January 2017.
  154. ^ Rizun, Peter R.; Wilmer, Christopher E.; Burley, Richard Ford; Miller, Andrew (2015). “How to Write and Format an Article for Ledger” (PDF). Ledger1 (1): 1–12. doi:10.5195/LEDGER.2015.1 (inactive 15 January 2021). ISSN 2379-5980OCLC 910895894Archived (PDF) from the original on 22 September 2015. Retrieved 11 January 2017. 

Further reading

External links

Wikiversity has learning resources about Blockchain
  •  Media related to Blockchain at Wikimedia Commons

Categories

” (WP)

Sources:

Fair Use Sources:

Categories
Data Science - Big Data Software Engineering

Doubly linked list

” (WP)

In computer science, a doubly linked list is a linked data structure that consists of a set of sequentially linked records called nodes. Each node contains three fields: two link fields (references to the previous and to the next node in the sequence of nodes) and one data field. The beginning and ending nodes’ previous and next links, respectively, point to some kind of terminator, typically a sentinel node or null, to facilitate traversal of the list. If there is only one sentinel node, then the list is circularly linked via the sentinel node. It can be conceptualized as two singly linked lists formed from the same data items, but in opposite sequential orders.

A doubly linked list whose nodes contain three fields: an integer value, the link to the next node, and the link to the previous node.
A doubly linked list whose nodes contain three fields: the link to the previous node, an integer value, and the link to the next node.

The two node links allow traversal of the list in either direction. While adding or removing a node in a doubly linked list requires changing more links than the same operations on a singly linked list, the operations are simpler and potentially more efficient (for nodes other than first nodes) because there is no need to keep track of the previous node during traversal or no need to traverse the list to find the previous node, so that its link can be modified.

Nomenclature and implementation

The first and last nodes of a doubly linked list are immediately accessible (i.e., accessible without traversal, and usually called head and tail) and therefore allow traversal of the list from the beginning or end of the list, respectively: e.g., traversing the list from beginning to end, or from end to beginning, in a search of the list for a node with specific data value. Any node of a doubly linked list, once obtained, can be used to begin a new traversal of the list, in either direction (towards beginning or end), from the given node.

The link fields of a doubly linked list node are often called next and previous or forward and backward. The references stored in the link fields are usually implemented as pointers, but (as in any linked data structure) they may also be address offsets or indices into an array where the nodes live.

Basic algorithms

Consider the following basic algorithms written in Ada:

Open doubly linked lists

record DoublyLinkedNode {
    next // A reference to the next node
    prev // A reference to the previous node
    data // Data or a reference to data
}
record DoublyLinkedList {
     DoublyLinkedNode firstNode   // points to first node of list
     DoublyLinkedNode lastNode    // points to last node of list
}

Traversing the list

Traversal of a doubly linked list can be in either direction. In fact, the direction of traversal can change many times, if desired. Traversal is often called iteration, but that choice of terminology is unfortunate, for iteration has well-defined semantics (e.g., in mathematics) which are not analogous to traversal.

Forwards

node  := list.firstNode
while node ≠ null
    <do something with node.data>
    node  := node.next

Backwards

node  := list.lastNode
while node ≠ null
    <do something with node.data>
    node  := node.prev

Inserting a node

These symmetric functions insert a node either after or before a given node:

function insertAfter(List list, Node node, Node newNode)
    newNode.prev  := node      
    if node.next == null
        newNode.next  := null -- (not always necessary)
        list.lastNode  := newNode
    else
        newNode.next  := node.next
        node.next.prev  := newNode
    node.next  := newNode
function insertBefore(List list, Node node, Node newNode)
    newNode.next  := node
    if node.prev == null
        newNode.prev  := null -- (not always necessary)
        list.firstNode  := newNode
    else
        newNode.prev  := node.prev
        node.prev.next  := newNode
    node.prev  := newNode

We also need a function to insert a node at the beginning of a possibly empty list:

function insertBeginning(List list, Node newNode)
    if list.firstNode == null
        list.firstNode  := newNode
        list.lastNode   := newNode
        newNode.prev  := null
        newNode.next  := null
    else
        insertBefore(list, list.firstNode, newNode)

A symmetric function inserts at the end:

function insertEnd(List list, Node newNode)
     if list.lastNode == null
         insertBeginning(list, newNode)
     else
         insertAfter(list, list.lastNode, newNode)

Removing a node

Removal of a node is easier than insertion, but requires special handling if the node to be removed is the firstNode or lastNode:

function remove(List list, Node node)
    if node.prev == null
        list.firstNode  := node.next
    else
        node.prev.next  := node.next
    if node.next == null
        list.lastNode  := node.prev
    else
        node.next.prev  := node.prev

One subtle consequence of the above procedure is that deleting the last node of a list sets both firstNode and lastNode to null, and so it handles removing the last node from a one-element list correctly. Notice that we also don’t need separate “removeBefore” or “removeAfter” methods, because in a doubly linked list we can just use “remove(node.prev)” or “remove(node.next)” where these are valid. This also assumes that the node being removed is guaranteed to exist. If the node does not exist in this list, then some error handling would be required.

Circular doubly linked lists

Traversing the list

Assuming that someNode is some node in a non-empty list, this code traverses through that list starting with someNode (any node will do):

Forwards

node  := someNode
do
    do something with node.value
    node  := node.next
while node ≠ someNode

Backwards

node  := someNode
do
    do something with node.value
    node  := node.prev
while node ≠ someNode

Notice the postponing of the test to the end of the loop. This is important for the case where the list contains only the single node someNode.

Inserting a node

This simple function inserts a node into a doubly linked circularly linked list after a given element:

function insertAfter(Node node, Node newNode)
    newNode.next  := node.next
    newNode.prev  := node
    node.next.prev  := newNode
    node.next       := newNode

To do an “insertBefore”, we can simply “insertAfter(node.prev, newNode)”.

Inserting an element in a possibly empty list requires a special function:

function insertEnd(List list, Node node)
    if list.lastNode == null
        node.prev := node
        node.next := node
    else
        insertAfter(list.lastNode, node)
    list.lastNode := node

To insert at the beginning we simply “insertAfter(list.lastNode, node)”.

Finally, removing a node must deal with the case where the list empties:

function remove(List list, Node node);
    if node.next == node
        list.lastNode := null
    else
        node.next.prev := node.prev
        node.prev.next := node.next
        if node == list.lastNode
            list.lastNode := node.prev;
    destroy node

Deleting a node

As in doubly linked lists, “removeAfter” and “removeBefore” can be implemented with “remove(list, node.prev)” and “remove(list, node.next)”.

Advanced concepts

Asymmetric doubly linked list

An asymmetric doubly linked list is somewhere between the singly linked list and the regular doubly linked list. It shares some features with the singly linked list (single-direction traversal) and others from the doubly linked list (ease of modification)

It is a list where each node’s previous link points not to the previous node, but to the link to itself. While this makes little difference between nodes (it just points to an offset within the previous node), it changes the head of the list: It allows the first node to modify the firstNode link easily.[1][2]

As long as a node is in a list, its previous link is never null.

Inserting a node

To insert a node before another, we change the link that pointed to the old node, using the prev link; then set the new node’s next link to point to the old node, and change that node’s prev link accordingly.

function insertBefore(Node node, Node newNode)
    if node.prev == null
        error "The node is not in a list"
    newNode.prev  := node.prev
    atAddress(newNode.prev)  := newNode
    newNode.next  := node
    node.prev = addressOf(newNode.next)
function insertAfter(Node node, Node newNode)
    newNode.next  := node.next
    if newNode.next != null
        newNode.next.prev = addressOf(newNode.next)
    node.next  := newNode
    newNode.prev  := addressOf(node.next)

Deleting a node

To remove a node, we simply modify the link pointed by prev, regardless of whether the node was the first one of the list.

function remove(Node node)
    atAddress(node.prev)  := node.next
    if node.next != null
        node.next.prev = node.prev
    destroy node

See also

References

  1. ^ http://www.codeofhonor.com/blog/avoiding-game-crashes-related-to-linked-lists
  2. ^ https://github.com/webcoyote/coho/blob/master/Base/List.h

Categories

” (WP)

Sources:

Fair Use Sources:

Categories
Data Science - Big Data Software Engineering

Linked data structure

” (WP)

In computer science, a linked data structure is a data structure which consists of a set of data records (nodes) linked together and organized by references (links or pointers). The link between data can also be called a connector.

In linked data structures, the links are usually treated as special data types that can only be dereferenced or compared for equality. Linked data structures are thus contrasted with arrays and other data structures that require performing arithmetic operations on pointers. This distinction holds even when the nodes are actually implemented as elements of a single array, and the references are actually array indices: as long as no arithmetic is done on those indices, the data structure is essentially a linked one.

Linking can be done in two ways – using dynamic allocation and using array index linking.

Linked data structures include linked listssearch treesexpression trees, and many other widely used data structures. They are also key building blocks for many efficient algorithms, such as topological sort[1] and set union-find.[2]

Common types of linked data structures

Linked lists

A linked list is a collection of structures ordered not by their physical placement in memory but by logical links that are stored as part of the data in the structure itself. It is not necessary that it should be stored in the adjacent memory locations. Every structure has a data field and an address field. The Address field contains the address of its successor.

Linked list can be singly, doubly or multiply linked and can either be linear or circular.Basic properties

  • Objects, called nodes, are linked in a linear sequence.
  • A reference to the first node of the list is always kept. This is called the ‘head’ or ‘front’.[3]


A linked list with three nodes contain two fields each: an integer value and a link to the next nodeA linked list with a single node.

Example in Java

This is an example of the node class used to store integers in a Java implementation of a linked list:

public class IntNode {
     public int value;
     public IntNode link;
     public IntNode(int v) { value = v; }
}

Example in C

This is an example of the structure used for implementation of linked list in C:

struct node
{
	int val;
	struct node *next;
};

This is an example using typedefs:

typedef struct node node;

struct node
{
	int val;
	node *next;
};

Note: A structure like this which contains a member that points to the same structure is called a self-referential structure.

Example in C++

This is an example of the node class structure used for implementation of linked list in C++:

class Node
{
	int val;
	Node *next;
};

Search trees

A search tree is a tree data structure in whose nodes data values can be stored from some ordered set, which is such that in an in-order traversal of the tree the nodes are visited in ascending order of the stored values.Basic properties

  • Objects, called nodes, are stored in an ordered set.
  • In-order traversal provides an ascending readout of the data in the tree.

Advantages and disadvantages

Linked list versus arrays

Compared to arrays, linked data structures allow more flexibility in organizing the data and in allocating space for it. In arrays, the size of the array must be specified precisely at the beginning, which can be a potential waste of memory, or an arbitrary limitation which would later hinder functionality in some way. A linked data structure is built dynamically and never needs to be bigger than the program requires. It also requires no guessing at creation time, in terms of how much space must be allocated. This is a feature that is key in avoiding wastes of memory.

In an array, the array elements have to be in a contiguous (connected and sequential) portion of memory. But in a linked data structure, the reference to each node gives users the information needed to find the next one. The nodes of a linked data structure can also be moved individually to different locations within physical memory without affecting the logical connections between them, unlike arrays. With due care, a certain process or thread can add or delete nodes in one part of a data structure even while other processes or threads are working on other parts.

On the other hand, access to any particular node in a linked data structure requires following a chain of references that are stored in each node. If the structure has n nodes, and each node contains at most b links, there will be some nodes that cannot be reached in less than logb n steps, slowing down the process of accessing these nodes – this sometimes represents a considerable slowdown, especially in the case of structures containing large numbers of nodes. For many structures, some nodes may require worst case up to n−1 steps. In contrast, many array data structures allow access to any element with a constant number of operations, independent of the number of entries.

Broadly the implementation of these linked data structure is through dynamic data structures. It gives us the chance to use particular space again. Memory can be utilized more efficiently by using these data structures. Memory is allocated as per the need and when memory is not further needed, deallocation is done.

General disadvantages

Linked data structures may also incur in substantial memory allocation overhead (if nodes are allocated individually) and frustrate memory paging and processor caching algorithms (since they generally have poor locality of reference). In some cases, linked data structures may also use more memory (for the link fields) than competing array structures. This is because linked data structures are not contiguous. Instances of data can be found all over in memory, unlike arrays.

In arrays, nth element can be accessed immediately, while in a linked data structure we have to follow multiple pointers so element access time varies according to where in the structure the element is.

In some theoretical models of computation that enforce the constraints of linked structures, such as the pointer machine, many problems require more steps than in the unconstrained random access machine model.

See also

References

  1. ^ Donald KnuthThe Art of Computer Programming
  2. ^ Bernard A. Galler and Michael J. Fischer. An improved equivalence algorithm. Communications of the ACM, Volume 7, Issue 5 (May 1964), pages 301–303. The paper originating disjoint-set forests. ACM Digital Library
  3. ^ http://www.cs.toronto.edu/~hojjat/148s07/lectures/week5/07linked.pdf

Categories

” (WP)

Sources:

Fair Use Sources:

Categories
C Language C# .NET C++ Cloud Data Science - Big Data DevOps Django Web Framework Flask Web Framework Go Programming Language Java JavaScript Kotlin PowerShell Python Ruby Software Engineering Spring Framework Swift TypeScript

Integrated Development Environment (IDE)

“An integrated development environment (IDE) is a software application that provides comprehensive facilities to computer programmers for software development. An IDE normally consists of at least a source code editorbuild automation tools and a debugger. Some IDEs, such as Visual Studio, NetBeans and Eclipse, contain the necessary compilerinterpreter, or both; others, such as SharpDevelop and Lazarus, do not.” (WP)

“The boundary between an IDE and other parts of the broader software development environment is not well-defined; sometimes a version control system or various tools to simplify the construction of a graphical user interface (GUI) are integrated. Many modern IDEs also have a class browser, an object browser, and a class hierarchy diagram for use in object-oriented software development.” (WP)

Categories
Bibliography

Techopedia.com

Fair Use Source: techopedia.com (TcPd)

Categories
C# .NET Cloud Data Science - Big Data History Software Engineering

Microsoft SQL Server

“” (CS9PR)

Previous: — Next:

Sources:

Fair Use Sources: