Affiliation:
1. Mathematical Science Center, University of Yamanashi, Kofu, Yamanashi 400-8511, Japan k-masaki@yamanashi.ac.jp
Abstract
Multistate Hopfield models, such as complex-valued Hopfield neural networks (CHNNs), have been used as multistate neural associative memories. Quaternion-valued Hopfield neural networks (QHNNs) reduce the number of weight parameters of CHNNs. The CHNNs and QHNNs have weak noise tolerance by the inherent property of rotational invariance. Klein Hopfield neural networks (KHNNs) improve the noise tolerance by resolving rotational invariance. However, the KHNNs have another disadvantage of self-feedback, a major factor of deterioration in noise tolerance. In this work, the stability conditions of KHNNs are extended. Moreover, the projection rule for KHNNs is modified using the extended conditions. The proposed projection rule improves the noise tolerance by a reduction in self-feedback. Computer simulations support that the proposed projection rule improves the noise tolerance of KHNNs.
Subject
Cognitive Neuroscience,Arts and Humanities (miscellaneous)
Reference47 articles.
1. Multivalued threshold functions. I. Boolean complex-threshold functions and their generalization;Aizenberg;Cybernetics and Systems Analysis,1971
2. Multivalued threshold functions. II. Synthesis of multivalued threshold elements;Aizenberg;Cybernetics and Systems Analysis,1973
3. Characteristics of sparsely encoded associative memory;Amari;Neural Networks,1989
4. A complex-valued neuron to transform gray level images to phase information;Aoki;Proceedings of the International Conference on Neural Information Processing,2002
5. An image storage system using complex-valued associative memories;Aoki;Proceedings of the International Conference on Pattern Recognition,2000