Hostname: page-component-586b7cd67f-gb8f7 Total loading time: 0 Render date: 2024-12-04T09:26:43.020Z Has data issue: false hasContentIssue false
  • English
  • Français

Philosophy and Machine Learning

Published online by Cambridge University Press:  01 January 2020

Paul Thagard
Paul Thagard*
Affiliation:
Cognitive Science Laboratory, Princeton University, Princeton, NJ 08542, U.S.A.
Get access Rights & Permissions [Opens in a new window]

Extract

Philosophers since the ancient Greeks have investigated the nature of different kinds of inference. Although deductive inference in the form of Aristotelian syllogisms and Fregean formal logic has predominated, much attention has also been paid to induction, inference where the conclusion does not follow necessarily from the premises.

Type
Research Article
Information
Canadian Journal of Philosophy , Volume 20 , Issue 2 , June 1990 , pp. 261 - 276
Copyright
Copyright © The Authors 1990

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

* I am grateful to Mohan Matthen for suggesting this review paper, and to him, Gregory Nowak, and Gilbert Harman for valuable comments on a previous draft.

1 Holland, J., Holyoak, K., Nisbett, R., and Thagard, P., Induction: Processes of Inference, Learning, and Discovery (Cambridge, MA: M.I.T. Press/Bradford Books 1986)Google Scholar

2 For example, Minsky, M. (‘A Framework for Representing Knowledge,’ in Winston, P.H., ed., The Psychology of Computer Vision [New York: McGraw-Hill 1975] 211-77Google Scholar) advocates structures called frames. Some cognitive scientists contend that some knowledge is represented in images: see Kosslyn, S., Image and Mind (Cambridge: Harvard University Press 1980)Google Scholar. Some connectionists advocate distributed representations, discussed below.

3 An exception is Cheeseman, P., ‘In Defense of Probability,’ Proceedings of the Ninth International Joint Conference on Artificial Intelligence, vol. 2 (Los Altos: Morgan Kaufmann 1985) 1002-9.Google Scholar

4 Mitchell, T., Keller, R., and Kedar-Cabelli, S., ‘Explanation-based Generalization: A Unifying View,’ Machine Learning 1 (1986) 47–80CrossRefGoogle Scholar; DeJong, G. and Mooney, R., ‘Explanation-based Learning: An Alternative View,’ Machine Learning 1 (1986) 145-76CrossRefGoogle Scholar. AI researchers are generally vague about what an explanation is and could profit from attention to the philosophical literature on the subject. For most of them, explanation is essentially deduction.

5 See the reference to Minsky's frames in note 2 above. Further defense of this view of concepts as complex structures is in P. Thagard, ‘Concepts and Conceptual Change,’ forthcoming in Synthese.

6 Rumelhart, D., McClelland, J., and the PDP Research Group, Parallel Distributed Processing: Explorations in the Microstructure of Cognition, 2 vol. (Cambridge, MA: M.I.T. Press/Bradford Books 1986)CrossRefGoogle Scholar; G. Hinton, ‘Connectionist Learning Procedures,’ Artificial Intelligence, forthcoming. The last few years of the Proceedings of the annual meetings of the Cognitive Science Society, published by Lawrence Erlbaum, contain a good sampling of connectionist work. Canadian connectionists are known as ‘cannuctionists.'

7 Sejnowsky, T. and Rosenberg, C., ‘Parellel Networks that Learn to Pronounce English Text,’ Complex Systems 1 (1987) 145-68Google Scholar

8 See for example the papers by Jerry Fodor and others in Pinker, S. and Mehler, J., eds., Connections and Symbols (Cambridge, MA: M.I.T. Press 1988)CrossRefGoogle Scholar and Smolensky, Paul, ‘On the Proper Treatment of Connectionism,’ Behavioral and Brain Sciences 11 (1988) 1–23CrossRefGoogle Scholar. The latter article is followed by more than 30 commentaries.

9 Thagard, P., ‘Explanatory Coherence,’ Behavioral and Brain Sciences 12 (1989), 435-67CrossRefGoogle Scholar; Thagard, P. and Nowack, G., ‘The Explanatory Coherence of Continental Drift,’ in Fine, A. and Leplin, J., eds., PSA 1988, vol. 1 (East Lansing, MI: Philosophy of Science Association 1988) 118-26Google Scholar

10 Two new volumes provide the best overview of work on analogy: Helman, D., ed., Analogical Reasoning (Dordrecht: Kluwer 1988)CrossRefGoogle Scholar; Vosniadou, S. and Ortony, A., eds., Similarity and Analogy (Cambridge: Cambridge University Press 1989)Google Scholar. For Gentner's approach, see her paper in the latter volume and B. Falkenhainer, K. Forbus, and D. Gentner, ‘The Structure-Mapping Engine,’ Artificial Intelligence, in press. For a connectionist alternative, see Holyoak, K. and Thagard, P., ‘Analogical Mapping by Constraint Satisfaction,’ Cognitive Science 13 (1989) 295–355.CrossRefGoogle Scholar

11 Reggia, J., Nau, D., and Wang, P., ‘Diagnostic Expert Systems Based on a Setcovering Model,’ International Journal of Man-Machine Studies 19 (1983) 437-60CrossRefGoogle Scholar; Josephson, J., Chandrasekaran, B., Smith, J., and Tanner, M., ‘A Mechanism for Forming Composite Explanatory Hypotheses,’ IEEE Transactions on Systems, Man, and Cybernetics (1987) 445-54CrossRefGoogle Scholar. Diagnosis can be understand as a kind of inference to the best explanation, using the term due to Harman, G., Thought (Princeton: Princeton University Press 1973).Google Scholar

12 Dreyfus, H., What Computers Can't Do, 2nd edn. (New York: Harper 1979)Google Scholar

13 Thagard, P., ‘Parallel Computation and the Mind-body Problem,’ Cognitive Science 10 (1986) 301-18CrossRefGoogle Scholar

14 Buchanan, B., ‘Aritificial Intelligence as an Experimental Science,’ in Fetzer, J., ed., Aspects of Artificial Intelligence (Dordrecht: Kluwer 1988) 209-50CrossRefGoogle Scholar

15 Langley, P., Simon, H., Bradshaw, G., and Zytkow, J., Scientific Discovery (Cambridge, MA: M.I.T. Press/Bradford Books 1987)CrossRefGoogle Scholar; Kulkarni, D. and Simon, H., ‘The Processes of Scientific Discovery,’ Cognitive Science 12 (1988) 139-77Google Scholar; Thagard, P., Computational Philosophy of Science (Cambridge, MA: M.I.T. Press/Bradford Books 1988)CrossRefGoogle Scholar; J. Shrager and P. Langley, eds., Computational Approaches to Scientific Discovery (papers from the 1989 workshop), forthcoming.

16 Goldman, A., Epistemology and Cognition (Cambridge, MA: Harvard University Press 1986)Google Scholar

17 Harman, G., Change in View (Cambridge, MA: M.I.T. Press/Bradford Books 1986)Google Scholar; Lycan, W., Judgement and Justification (Cambridge: Cambridge University Press 1988)Google Scholar

18 P. Thagard, Computational Philosophy of Science; ‘Concepts and Conceptual Change'

19 Sloman, A., The Computer Revolution in Philosophy (Atlantic Highlands: Humanities Press 1978)Google Scholar

20 Glymour, C., Scheines, R., Spirtes, P., and Kelly, K., Discovering Causal Structure (Orlando: Academic Press 1987)Google Scholar

21 Darden, L. and Rada, R., ‘Hypothesis Formation Using Part-Whole Interrelations,’ in Hellman, D., ed., Analogical Reasoning (Dordrecht: Reidel 1988), 341-75CrossRefGoogle Scholar

22 G. Harman, unpublished work. See also Change in View (Cambridge, MA: M.I.T. Press/Bradford Books 1986).

23 Pollock, John, ‘Defeasible Reasoning,’ Cognitive Science 11 (1987) 481–518CrossRefGoogle Scholar; Nute, D., ‘Defeasible Reasoning: A Philosophical Analysis in Prolog,’ in Fetzer, J., eds., Aspects of Artificial Intelligence (Dordrecht: Kluwer 1988) 133-61Google Scholar; Horty, J. and Thomason, R., ‘Mixing Strict and Defeasible Inheritance,’ Proceedings of the National Conference on Artificial Intelligence (San Mateo: Morgan Kaufmann 1988)Google Scholar

24 Kelly, K., ‘Theory Discovery and the Hypothesis Language,’ in Laird, J., ed., Proceedings of the Fifth International Conference 011 Machine Learning (San Mateo: Morgan Kaufmann 1988) 325-38Google Scholar

25 The best introduction to LISP I know is Anderson, J., Corbett, A., and Reiser, B., Essential LISP (Reading, MA: Addison-Wesley 1986)Google Scholar. It's been said that a philosopher of cognitive science who doesn't know LISP is like a philosopher of physics who doesn't know calculus.

26 The standard reference on PROLOG is Clocksin, W., and Mellish, C., Programming in Prolog (Berlin: Springer-Verlag 1981)Google Scholar.

27 Dummett, M., Truth and Other Enigmas (Cambridge, MA: Harvard University Press 1978) 441Google Scholar

28 A broad picture of current cognitive science can be got by perusing the journal Cognitive Science or the Proceedings of the Annual Conference of the Cognitive Science Society, published by Morgan Kaufmann. The best integrated introduction is Johnson-Laird, P., The Computer and the Mind (Cambridge, MA: Morgan Kaufmann 1988)Google Scholar.