Adaptive architectures for Hebbian network models

(Received 13 December 1991, accepted in final form 24 January 1992)

Abstract
We present a novel produce which allows a neural network to evolve to a quasi-optimal connectivity such that information to be memorized is engraved as efficiently as possible. The emerging connectivity structure of the resulting partially connected network depends strongly on the information the network is asked to memorize. A stability measure of the quality of storage as a function of the degree of the connectivity is shown to attain a maximum value, which lies substantially above the stability of the traditional fully connected system.