Neural networks are a special class of distributed active systems that consist of discrete two-state elements linked by long-range activatory and inhibitory connections. The applicability of these formal models to description of the actual neural net of the brain remains doubtful. They give what is probably no more than a rough sketch of the extremely complex processes involved in the operation of the brain. Nevertheless, one can easily construct artificial active networks with these properties. Furthermore, artificial neural networks can be used to perform analog information processing. To use them in this way, one needs to be able to engineer the networks to give them the desired activity patterns.
KeywordsBack Propagation Spin Glass Input Pattern Associative Memory Hide Unit
Unable to display preview. Download preview PDF.
- 6.5D.C. Hebb: The Organization of Behavior: A Neuropsycological Theory (Wiley, New York 1957)Google Scholar
- 6.10W. Kmzel: Physica Scripta 39, 171–194 (1988)Google Scholar
- 6.13E. Gardner, N. Stroud, DJ. Wallace: Edinburgh Preprint 87/394Google Scholar
- 6.16M. Opper: Phys. Rev. Lett. 22, 235 (1988)Google Scholar
- 6.21J J. Hopfield, D.W. Tank: “Collective computation with continuous variables”, in Disordered Systems and Biological Organization, ed. by E. Bienenstock et al. (Springer, Berlin, Heidelberg 1986)Google Scholar
- 6.28F. Rosenblatt:Principles of Neurodynamics (Spartan Books, New York 1959)Google Scholar
- 6.29D.E. Rumelhart, J.L. McClelland: “On learning the past tenses of English verbs”, in Parallel Distributed Processing, Vol.2, ed. by D.E. Rumelhart et.al. (MIT Press, Cambridge, MA 1986), pp. 216–271Google Scholar
- 6.32D.E. Rumelhart, G.E. Hinton, RJ. Williams: “Learning internal representations by error propagation”, in Parallel Distributed Processing, Vol.1, ed. by D.E. Rumelhart et.al. (MIT Press, Cambridge, MA 1986), pp. 318–362Google Scholar
- 6.33Y. Le Cun: “A learning scheme for asymmetric threshold networks”, in Proc. Int. Conf. Cognitiva 85 (Paris, 1985) pp. 599–604Google Scholar
- 6.39G.E. Hinton, T J. Sejnowski: “Optimal perceptual inference”, in Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Washington, 1983) pp. 448–453Google Scholar
- 6.40G.E. Hinton, TJ. Sejnowski: “Learning and releaming in Boltzmann machines”, in Parallel Distributed Processing, Vol. 1, ed. by D.E. Rumelhart et al. (MIT Press, Cambridge, MA 1986) pp. 282–317Google Scholar
- 6.52A.S. Mikhailov, I.V. Mitkov, N.A. Sveshnikov: J. Nonlinear Biol. 1 (1990), in pressGoogle Scholar