Fontanari, J. F. and Meir, R. (1989) Learning noisy patterns in a Hopfield network. Physical Review A, 40 (5). pp. 2806-2809. ISSN 0556-2791 http://resolver.caltech.edu/CaltechAUTHORS:FONpra89
See Usage Policy.
Use this Persistent URL to link to this item: http://resolver.caltech.edu/CaltechAUTHORS:FONpra89
We study the ability of a Hopfield network with a Hebbian learning rule to extract meaningful information from a noisy environment. We find that the network is able to learn an infinite number of ancestor patterns, having been exposed only to a finite number of noisy versions of each. We have also found that there is a regime where the network recognizes the ancestor patterns very well, while performing very poorly on the noisy patterns to which it had been exposed during the learning stage.
|Additional Information:||©1989 The American Physical Society Received 5 June 1989 The research at the California Institute of Technology was supported by Contract No. N00014-87-K-0377 from the U.S. Office of Naval Research. J.F.F. was partly supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico and R.M. is supported by the Weizmann Foundation.|
|Usage Policy:||No commercial reproduction, distribution, display or performance rights in this work are provided.|
|Deposited By:||Archive Administrator|
|Deposited On:||14 Dec 2006|
|Last Modified:||26 Dec 2012 09:23|
Repository Staff Only: item control page