Neural networks with many-neuron interactions

G.A. Kohring

The static and dynamical properties of neural networks having many-neuron interactions are studied analytically and numerically. The storage capacity of such networks is found to be unchanged from that of the more widely studied case of two-neuron interactions implying that these networks store information no more efficiently. The size of the basins of attraction in the many-neuron case is calculated exactly from a solution of the network dynamics at full connectivity and reveals that networks with many-neuron interactions are better at pattern discrimination than the simpler networks with only two-neuron interactions.

Mots clés: learning systems; neural nets; pattern recognition

Domaines: Articles anciens

G.A. Kohring. Neural networks with many-neuron interactions. Journal de Physique, 1990, 51 (2), pp.145-155. ⟨10.1051/jphys:01990005102014500⟩. ⟨jpa-00212356⟩