A Continuous-Time Hopfield Net Simulation of Discrete Neural Networks

0403862 - UIVT-O 20000040 RIV CA eng C - Konferenční příspěvek (zahraniční konf.)
Šíma, Jiří - Orponen, P.
A Continuous-Time Hopfield Net Simulation of Discrete Neural Networks.
Proceedings of the Second ICSC Symposium on Neural Computation. Wetaskiwin: ICSC Academic Press, 2000 - (Bothe, H.; Rojas, R.), s. 36-42. ISBN 3-906454-22-3.
[NC'2000. ICSC Symposium on Neural Computation /2./. Berlin (DE), 23.05.2000-26.05.2000]
Grant CEP: GA ČR GA201/98/0717; GA AV ČR IAB2030007
Výzkumný záměr: AV0Z1030915
Klíčová slova: neural networks * analog computation * computational power * continuous-time Hopfield
Kód oboru RIV: BA - Obecná matematika

We investigate the computational power of continuous-time symmetric Hopfield nets. As is well known, such networks have very constrained Liapunov-function controlled dynamics. Nevertheless, we show that they are universal and efficient computational devices, in the sense that any convergent fully parallel computation by a network of n discrete-time binary neurons, with in general asymmetric interconnections, can be simulated by a symmetric continuous-time Hopfield net containing only 14n+6 units...
Trvalý link: http://hdl.handle.net/11104/0124150