Discovery system (artificial intelligence)
A discovery system is an artificial intelligence system that attempts to discover new scientific concepts or laws. The aim of discovery systems is to automate scientific data analysis and the scientific discovery process. Ideally, an artificial intelligence system should be able to search systematically through the space of all possible hypotheses, and yield the hypothesis or set of equally likely hypotheses that best describe the complex patterns in data.[1][2]
During the era know knows as the second AI summer (approximately 1978-1987), various systems akin to the eras dominant expert systems were developed to tackle the problem of extracting scientific hypotheses from data, with or without interaction with a human scientist. These systems included Autoclass[3], Automated Mathematician[4][5], Eurisko[6], which aimed at general-purpose hypothesis discovery, and more specific systems such as Dalton, which uncovers molecular properties from data.
The dream of building systems that uncover scientific hypotheses was pushed to the background with the second AI winter, and the subsequent resurgence of subsymbolic methods such as neural networks. These methods emphasize prediction over explanation, and often yields models which works well but are difficult or impossible to explain, and have thus earned the name black box AI. A black-box model cannot be considered a scientific hypothesis, and this development has even led some researchers to suggest that the traditional aim of science - to uncover hypotheses and theories about the structure of reality is obsolete.[7][8] Other researchers disagree, and argue that subsymbolic methods are useful, perhaps just not for generating scientific theories[9][10][11].
Discovery systems form the 70s and 80s
- Autoclass was a Bayesian Classification System written in 1986[3]
- Automated Mathematician was one of the earliest successful discovery systems. It was written in 1977 and worked by generating a modifying small Lisp programs
- Eurisko was a Sequel to Automated Mathematician written in 1984
- Dalton is a still maintained program capable of calculating various molecular properties initially launched in 1983 and available in open source since 2017
- Glauber is a scientific discovery method written in the context of computational philosophy of science launched in 1983
See also
References
- ^ Shen, Wei-Min (1990). "Functional transformations in AI discovery systems". Artificial Intelligence. 41 (3). Elsevier BV: 257–272. doi:10.1016/0004-3702(90)90045-2. ISSN 0004-3702.
- ^ Gil, Yolanda; Greaves, Mark; Hendler, James; Hirsh, Haym (2014-10-10). "Amplify scientific discovery with artificial intelligence". Science. 346 (6206). American Association for the Advancement of Science (AAAS): 171–172. doi:10.1126/science.1259439. ISSN 0036-8075.
- ^ a b Cheeseman, PETER; Kelly, JAMES; Self, MATTHEW; Stutz, JOHN; Taylor, WILL; Freeman, DON (1988-01-01), Laird, John (ed.), "AutoClass: A Bayesian Classification System", Machine Learning Proceedings 1988, San Francisco (CA): Morgan Kaufmann, pp. 54–64, doi:10.1016/b978-0-934613-64-4.50011-6, ISBN 978-0-934613-64-4, retrieved 2022-07-24
- ^ Ritchie, G.D.; Hanna, F.K. (August 1984). "am: A case study in AI methodology". Artificial Intelligence. 23 (3): 249–268. doi:10.1016/0004-3702(84)90015-8.
- ^ Lenat, Douglas Bruce (1976). Am: An artificial intelligence approach to discovery in mathematics as heuristic search (Thesis).
- ^ Henderson, Harry (2007), "The Automated Mathematician", Artificial Intelligence: Mirrors for the Mind, Milestones in Discovery and Invention, Infobase Publishing, pp. 93–94, ISBN 9781604130591.
- ^ Anderson, Chris. "The End of Theory: The Data Deluge Makes the Scientific Method Obsolete". Wired. ISSN 1059-1028. Retrieved 2022-07-24.
- ^ Vutha, Amar. "Could machine learning mean the end of understanding in science?". The Conversation. Retrieved 2022-07-24.
- ^ Canca, Cansu (2018-08-28). "Machine Learning as the Enemy of Science? Not Really". Bill of Health. Retrieved 2022-07-24.
- ^ Wilstrup, Casper Skern (2022-01-30). "Are we replacing science with an AI oracle?". Medium. Retrieved 2022-07-24.
- ^ Christiansen, Michael; Wilstrup, Casper; Hedley, Paula L. (2022-06-28). "Explainable "white-box" machine learning is the way forward in preeclampsia screening". American Journal of Obstetrics & Gynecology. 0 (0). doi:10.1016/j.ajog.2022.06.057. ISSN 0002-9378. PMID 35779588.
External links
 | This artificial intelligence-related article is a stub. You can help Wikipedia by expanding it. |
 | This mathematics-related article is a stub. You can help Wikipedia by expanding it. |