The Enhanced Storage Capacity in Neural Networks with Low Activity Level

and

Published under licence by IOP Publishing Ltd
, , Citation M. V. Tsodyks and M. V. Feigel'man 1988 EPL 6 101 DOI 10.1209/0295-5075/6/2/002

0295-5075/6/2/101

Abstract

The modified Hopfield model defined in terms of "V-variables" (V = 0; 1), which is appropriate for storage of correlated patterns, is considered. The learning algorithm is proposed to enhance significantly the storage capacity in comparison with previous estimates. At low levels of neural activity, p ≪ 1, we obtain αc(p) ∼ (p|ln p|)-1 which resembles Gardner's estimate for the maximum storage capacity.

Export citation and abstract BibTeX RIS