Dynamic complex-valued associative memory with strong bias terms

Authors:
Yozo Suzuki;Michimasa Kitahara;Masaki Kobayashi
Affiliations:
Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, Kofu, Yamanashi, Japan;Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, Kofu, Yamanashi, Japan;Interdisciplinary Graduate School of Medicine and Engineering, University of Yamanashi, Kofu, Yamanashi, Japan
Venue:
ICONIP'11 Proceedings of the 18th international conference on Neural Information Processing - Volume Part I
Year:
2011

Citing 4
Cited 0

Lending direction to neural networks

Neural Networks
Complex-valued multistate neural associative memory

IEEE Transactions on Neural Networks
A new design method for the complex-valued multistate Hopfield associative memory

IEEE Transactions on Neural Networks
Improvements of Complex-Valued Hopfield Associative Memory by Using Generalized Projection Rules

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Complex-valued associative memory (CAM) can store multi-level patterns. Dynamic complex-valued associative memory (DCAM) can recall all stored patterns. The CAM stores the rotated patterns, which are typical spurious states, in addition to given training patterns. So DCAM recalls all the rotated patterns in the recall process. We introduce strong bias terms to avoid recalling the rotated patterns. By computer simulations, we can see that strong bias terms can avoid recalling the rotated patterns unlike simple bias terms.