New Computer Algorithm Enabling Simulation of Recall Process in Sparserly Encoded Hopfield-Like Neural Nretwork of Extremely Large Size

0404148 - UIVT-O 20000243 RIV SK eng C - Konferenční příspěvek (zahraniční konf.)
Frolov, A. A. - Húsek, Dušan - Snášel, V.
New Computer Algorithm Enabling Simulation of Recall Process in Sparserly Encoded Hopfield-Like Neural Nretwork of Extremely Large Size.
ALGORITMY 2000. Bratislava: Vydavateĺstvo STU, 2000 - (Handlovičová, A.; Komorníková, M.; Mikula, K.; Ševčovič, D.), s. 314-319. ISBN 80-227-1391-0.
[ALGORITMY 2000. Conference on Scientific Computing /15./. Podbanské (SK), 10.09.2000-15.09.2000]
Grant CEP: GA ČR GA201/00/1031
Výzkumný záměr: AV0Z1030915
Klíčová slova: Hopfield-like neural network * Hebbian rule
Kód oboru RIV: BA - Obecná matematika

The algorithm which minimazes the required computer memory to simulate neurodynamics in sparsely encoded Hopfield-like autoassociative memory is proposed. It does not require to keep in memory both connection matrix and a set of stored prototypes. The algorithm allows to simulate the recall process in the neural network of extremely large size and thus to calculate its asymptotic informational and dynamic properties when the number of network's nodes become infinite.
Trvalý link: http://hdl.handle.net/11104/0124416