Learning Algorithms — theory and practice
Computational Learning Theory is a recently-developed branch of mathematics which provides a framework for the discussion of experiments with learning machines, such as artificial neural networks. The basic ideas of the theory are described, and applied to an experiment involving the comparison of two learning machines. The experiment was devised so that the results could also be compared with those achieved by a human subject, and this comparison raises some interesting questions.
KeywordsLearning Algorithm Training Sample Target Concept Hypothesis Space Computational Learn Theory
Unable to display preview. Download preview PDF.
- P.L. Bartlett and R.C. Williamson, Investigating the distribution assumptions in the pac model, Proceedings of the 1991 Workshop on Computational Learning Theory, Morgan Kauffman, San Mateo, 1991.Google Scholar
- E. B. Baum, The perceptron algorithm is fast for non-malicious distributions, Neural Computation, 2 (1990) 249–261.Google Scholar
- R. L. Gregory and J. G. Wallace, Recovery from early blindness: a case study, Exp. Psychol. Soc. Monographs No. 2, Cambridge 1963.Google Scholar
- N. Littlestone, Learning quickly when irrelevant attributes abound, Machine Learning, 2 (1988) 245–318.Google Scholar
- F. Rosenblatt, Principles of Neurodynamics, Spartan Books, 1962.Google Scholar
- L. G. Valiant, A view of computational learning theory, NEC Research Symposium: Computation and Cognition (ed. C. W. Gear), SIAM, 1991.Google Scholar