Abstract
We will consider the dynamics of attractor neural networks with sign-constrained weights. In the presense of sign-constraints on weights three types of attractor can affect the dynamics: retrieval attractors, spurious attractors and uniform attractors. The uniform attracting states can dominate the dynamics if there is a substantial weight-sign bias. We will show that it is possible to define dynamic thresholds for a variety of learning rules which can eliminate uniform attracting states for any value of the weight-sign bias.
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Preview
Unable to display preview. Download preview PDF.
References
K.Y.M. Wong and C. Campbell, 1991, Competition between Attractors in Neural Networks with Sign Constrained Weights, J. Phys. A(to appear).
J-S. Jang, S-W. Jung, S-Y Lee and S-Y Shin, 1988, Optics Letters, 13 248.
C. Campbell and K.Y.M. Wong, 1990, Lecture Notes in Physics, 368 (Berlin: Springer-Verlag) 237.
D. J. Amit, K.Y.M. Wong and C. Campbell, 1989, J.Phys. A22, 2039.
A.D. Bruce, E.J. Gardner and D.J. Wallace, 1987, J.Phys. A20, 2909.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1991 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Campbell, C. (1991). Dynamic thresholds and attractor neural networks. In: Prieto, A. (eds) Artificial Neural Networks. IWANN 1991. Lecture Notes in Computer Science, vol 540. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0035894
Download citation
DOI: https://doi.org/10.1007/BFb0035894
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-54537-8
Online ISBN: 978-3-540-38460-1
eBook Packages: Springer Book Archive