On Informational Characteristics of Willshaw-Like Auto-Associative Memory

0404653 - UIVT-O 20020154 RIV CZ eng J - Journal Article
Frolov, A. A. - Rachkovskij, D. - Húsek, Dušan
On Informational Characteristics of Willshaw-Like Auto-Associative Memory.
Neural Network World. Roč. 12, č. 2 (2002), s. 141-157. ISSN 1210-0552
R&D Projects: GA ČR GA201/01/1192; GA ČR GA201/00/1031
Institutional research plan: AV0Z1030915
Keywords : Willshaw-like * space encoding * attractor neural network * binary Hebbian synapses * computer simulations * constant * informational capacity * attraction basis size * single-step approximation * Gibson-Robinson approximation
Subject RIV: BA - General Mathematics

A sparsely encoded Willshaw-like attractor neural network based on the binary Hebbian synapses is investigated analytically and by computer simulations. A special inhibition mechanism which supports a constant number of active neurons at each time step is used. The informational capacity and the size of attraction basis are evaluated for the Single-Step and the Gibson-Robinson approximations, as well as for experimental results.
Permanent Link: http://hdl.handle.net/11104/0003455