Advertisement

The I in AI (or why there is still none)

Prof. Dr. Computer Science (Artificial Intelligence, Software Engineering), Co-Founder AGISI.org at Computer Science Dept., Berlin School of Economics and Law
Jun. 11, 2021
Advertisement

More Related Content

Advertisement

More from Dagmar Monett(20)

Advertisement

The I in AI (or why there is still none)

  1. The I in AI (or why there is still none) El Futuro Digital de las Infraestructuras y la Sociedad Universidad de Castilla-La Mancha June 9th, 2021 (online) @dmonett Prof. Dr. Dagmar Monett Berlin School of Economics and Law, and AGISI.org dagmar@monettdiaz.com http://monettdiaz.com
  2. 2 @dmonett How to cite this source: Monett, D. (2021). The I in AI (or why there is still none). Keynote at the Webinar “El Futuro Digital de las Infraestructuras y la Sociedad,” Universidad de Castilla-La Mancha, Spain, June 9th, 2021 [online]. Available at: https://www.slideshare.net/dmonett/monett-2021-uclm (Accessed: access date) Note: This is a slightly extended version of the original slides. Slides with the symbol, like this one, were added after the talk. They include additional related information.
  3. What is (machine) intelligence? @dmonett
  4. 4 Cortical Columns Greg Dunn, 2014 www.gregadunn.com @dmonett
  5. A very brief history of (machine) intelligence @dmonett
  6. The scientific study of intelligence originated in the 1870s ... … about 150 years ago. 6 1870s @dmonett
  7. “The intelligent being, animal or man, supplies its wants, preserves its life, improves its condition, only by the exact accordance of its present prevision and the near or even distant future” (Taine 1875). 7 Taine, H. (1870). De l’ Intelligence. Two volumes, Paris (engl. traduction by T. D. Have: 1875). First scientific definition of intelligence? 1870 Hippolyte Adolphe Taine (April 21, 1828 – March 5, 1893) 1870 … among other 500+ @dmonett
  8. [Intelligence is] the power of good responses from the point of view of truth or facts (Thorndike); the ability to carry on abstract thinking (Terman); having learned or ability to learn to adjust oneself to the environment (Colvin); the capacity for knowledge (Henmon); the capacity to acquire capacity (Woodrow). 8 As referred to in Lanz, P. (2000). The Concept of Intelligence in Psychology and Philosophy. In Cruse, H., Dean, J., and Ritter, H. (eds) Prerational Intelligence: Adaptive Behavior and Intelligent Systems Without Symbols and Logic, Vol. 1, 19-30, Springer. 5 (14) educational psychologists define intelligence 1921 @dmonett
  9. “… stimulated by the invention of modern computers. This inspired a flood of new ideas about how machines could do what only minds had done previously” (Minsky 1985) 9 Minsky, M. (1985). The Society of Mind. Simon and Schuster, New York. 1950s Research in AI started in the 1950s @dmonett
  10. “[Intelligence is] the ability to solve hard problems” (Minsky 1985). 10 AI definition Minsky, M. (1985). The Society of Mind. Simon and Schuster, New York. 1985 @dmonett
  11. [Intelligence is] an elusive concept (Estes); an illusory unified capacity (Horn); a cognitive proficiency (Glaser); a polymorphous set of qualities elusive to define, explain, and measure (Brown); a pluralistic (Anastasi), context-dependent con- cept (Anastasi; Sternberg); a medley of important events, a mixture of different things (Horn); a finite set of independent abilities operating as a complex system (Detterman); the sum total of all cognitive processes (Das); a collective term for demonstrated, mental individual differences (Hunt); mental self-government (Sternberg); a judgement or attribution that people do, and not a quality residing in the individual (Goodnow); a hypothetical (Zigler), culture-bound, ethnocentric, and excessively narrow (Berry), societal construct, a concept in the mind of a society at large (Carroll). 11 A summary of some of the definitions that are included in Sternberg, R. J. and Detterman, D. K. (1986). What is intelligence? Contemporary Viewpoints on its Nature and Definition. Norwood, NJ: Ablex. 16 (25) leading psychologists define intelligence 1986 @dmonett
  12. “Artificial Intelligence is . . . the study of the computations that make it possible to perceive, reason, and act” (Winston 1992). 12 AI definition Winston, P. H. (1992). Artificial Intelligence. Third Edition, Addison-Wesley Publishing Company. 1992 @dmonett
  13. 13 (Monett et al., 2020) Monett, D., Lampe, N., Ehrlicher-Schmidt, M., & Bewer, N. (2020). Intelligence Catalog-guided Tracking of the Evolution of (machine) Intelligence: Preliminary results. In Basile, P., Basile, V., Cabrio, E., & Croce, D. (eds.), Proceedings of the 4th Workshop on Natural Language for Artificial Intelligence, NL4AI 2020, 2735: 118-129, CEUR-WS, co-located with the 19th International Conference of the Italian Association for Artificial Intelligence, AIxIA 2020, November 25th-27th, 2020, online event. Evolution of perceive, reason, and act(*) *: Examples of intelligence capabilities used in the texts of 9879 IJCAI papers (1997-2020). @dmonett
  14. “Intelligence is the capacity of a system to adapt to its environment while operating with insufficient knowledge and resources” (Wang 1995). 14 Wang, P. (1995). Non-Axiomatic Reasoning System: Exploring the Essence of Intelligence. Doctoral dissertation. Indiana University, Indianapolis, USA. AI definition 1995 @dmonett
  15. 15 A widely accepted definition of intelligence Gottfredson, L. S. (1997a). Mainstream science on intelligence: An editorial with 52 signatories, history, and bibliography. Intelligence, 24: 13-23. As cited in Haier, R. J. (2017). The Neuroscience of Intelligence. New York: Cambridge University Press. 1997 @dmonett
  16. “[Artificial Intelligence is] the science and engineering of making intelligent machines. ... It is related to the similar task of using computers to understand human intelligence” (McCarthy 2007). “Intelligence measures an agent’s ability to achieve goals in a wide range of environments” (Legg & Hutter 2007). 16 AI definitions McCarthy, J. (2007). What is Artificial Intelligence? Computer Science Department, School of Engineering, Stanford University. Legg, S. and Hutter, M. (2007). Universal Intelligence: A Definition of Machine Intelligence. Minds and Machines, 17(4):391-444, Springer. 2007 @dmonett
  17. A collection of 70+ definitions of intelligence. Both of human and machine intelligence. 17 Legg, S. and Hutter, M. (2007a). A Collection of Definitions of Intelligence. In B. Goertzel and P. Wang (eds.), Advances in Artificial General Intelligence: Concepts, Architectures and Algorithms, 157:17-24, IOS Press, UK. 2007 @dmonett
  18. “Artificial intelligence is that activity devoted to making machines intelligent, and intelligence is that quality that enables an entity to function appropriately and with foresight in its environment” (Nilsson 2010). “Intelligence is concerned mainly with rational action. Ideally, an intelligent agent takes the best possible action in a situation” (Russell & Norvig 2010). 18 Nilsson, N. J. (2010). The Quest for Artificial Intelligence: A History of Ideas and Achievements. Cambridge, UK: Cambridge University Press. Russell, S. J. and Norvig, P. (2010). Artificial Intelligence: A Modern Approach, Third Edition. Prentice Hall. AI definitions 2010 @dmonett
  19. There is no consensus @dmonett
  20. “There is very great disagreement concerning the concept of intelligence” (Journal editors 1921). “[A] substantial disagreement on a single definition still abounds” (Detterman 1986). “It is a testimony to the immaturity of our field that the question of what we mean when we talk about intelligence still doesn’t have a satisfying answer” (Chollet 2019) 20 Journal editors (1921). Intelligence and Its Measurement: A Symposium. Journal of Educational Psychology, Vol 12(3), 123- 147. Detterman, D. K. (1986). Qualitative Integration: The Last Word? In R. J. Sternberg and D. K. Detterman (eds.), What is intelligence? Contemporary Viewpoints on its Nature and Definition, pp. 163-166. Norwood, NJ: Ablex. Chollet, F. (2019). The Measure of Intelligence. arXiv:1911.01547 [cs.AI]. No consensus definition (i) @dmonett
  21. 21 “[The concept of artificial intelligence] is aspirational, a moving target based on those capabilities that humans possess but which machines do not” (Lipton 2018). “[I]t shifts with technological advances and our expectations from computers. That’s why it’s pretty hard to determine what is or isn’t AI” (Dickson 2019). No consensus definition (ii) Lipton, Z. C. (2018). From AI to ML to AI: On Swirling Nomenclature & Slurried Thought. Approximately Correct: Technical and Social Perspectives on Machine Learning. http://approximatelycorrect.com/ Dickson, B. (2019). 5 European companies that are (really) advancing AI. The Next Web. https://thenextweb.com/ ● Still no consensus definition of (A)I ● Very polarized concept ● Interdisciplinarity, different contexts and applications ● Misleading news and hype around AI damaging the field ● Need to know the boundaries of the discourse @dmonett
  22. 18 definitions of (human/machine) intelligence to agree upon in a survey. 340+ new, suggested definitions! 22 Monett, D. and Lewis, C. W. P. (2018). Getting clarity by defining Artificial Intelligence—A Survey. In Müller, Vincent C. (Ed.), Philosophy and Theory of Artificial Intelligence 2017. SAPERE 44 (pp. 212-214). Berlin: Springer. 2017-2019 @dmonett AGISI survey
  23. AGISI survey 14 Educational Psychology USA, Europe Prediction of behavior 14 Only definitions of human intelligence 25 Diverse Psychologies (educational, cognitive, behavioural, social, cross- cultural, etc.) USA, Europe Understanding of behavior 25 Mostly definitions of human intelligence 567 (academia: 79.7%) Computer Science, Engineering, Biology, Neurosciences, Philosophy, Cognitive Science, etc. 57+ countries Computation of behaviour 343 (+ 4128 opinions) Explicit distinction human vs. machine intelligence 23 A comparison Participants Primary affiliation Countries Focus Definitions Type of definitions Symposium Symposium 1921 1986 2019 @dmonett
  24. A well-defined definition of (machine) intelligence supports understanding, shapes the discourse, and advances the field. But which one? @dmonett
  25. 25 Still a widely accepted definition of intelligence Gottfredson, L. S. (1997a). Mainstream science on intelligence: An editorial with 52 signatories, history, and bibliography. Intelligence, 24: 13-23. As cited in Haier, R. J. (2017). The Neuroscience of Intelligence. New York: Cambridge University Press. 1997 Most accepted definition, AGISI survey @dmonett
  26. 1 target article on defining AI (Wang, 2019) 20 commentaries from leading AI experts 1 extended answer from target author 26 @dmonett 2020 Wang, P. (2019). On Defining Artificial Intelligence. Journal of Artificial General Intelligence, 10(2), 1–37. Monett, D., Lewis, C. W. P., & Thórisson, K. R. (Hrsg.) (2020). Special Issue “On Defining Artificial Intelligence.” Journal of Artificial General Intelligence, 11(2), 1–100. 2019 But still no general consensus in the AI community!
  27. The long-term dream of AI @dmonett
  28. “The long-term dream of AI is to build machines that have the full range of capabilities for intelligent actions that people have—to build machines that are self- aware, conscious and autonomous in the same way that people like you and me are. [...] The reality of AI for the foreseeable future is very different to the grand dream.” (Wooldridge 2020). 28 Wooldridge, M. (2020). The Road to Conscious Machines: The Story of AI. UK: Pelican Random House. 2020 @dmonett
  29. Dagmar Monett is Professor of Computer Science (Artificial Intelligence, Software Engineering) at the HWR Berlin. Shehas over 30 years of research and teaching experience. Co-founder of the AGI Sentinel Initiative, AGISI.org, dedicated to understanding intelligence in order to build beneficial AI. Other research areas include Machine Learning, Digital Education Ethics, and Computer Science Education. 29 A dose of reality! @dmonett Dagmar Monett es profesor de Ciencias de la Computación (Inteligencia Artificial, Ingeniería de Software) en el HWR de Berlín. Tiene más de 30 años de experiencia en investigación y docencia. Cofundador de la Iniciativa AGI Sentinel, AGISI.org, dedicada a comprender la inteligencia para construir una IA beneficiosa. Otras áreas de investigación incluyen aprendizaje automático, ética de la educación digital y educación en ciencias de la computación. (Google Translator 2021).
  30. 30 AI’s perception and PR problems @dmonett
  31. 31 Misleading news and AI hype Do not feed the hype! @dmonett
  32. 32 Needed, among others: - Multidisciplinarity - AI literacy - AI ethics - Understanding @dmonett
  33. 33 “[C]ellular structure of the rodent hippocampus, a region of the brain involved in learning, memory, and sensory integration.” Brainbow Hippocampus Greg Dunn and Brian Edwards, 2014, www.gregadunn.com A complex reality
  34. The I in AI (or why there is still none) Prof. Dr. Dagmar Monett Berlin School of Economics and Law, and AGISI.org dagmar@monettdiaz.com http://monettdiaz.com El Futuro Digital de las Infraestructuras y la Sociedad Universidad de Castilla-La Mancha June 9th, 2021 (online) @dmonett
Advertisement