Algorithm for optimizing bipolar interconnection weights with applicationsin associative memories and multitarget classification

An algorithm for optimizing a bipolar interconnection weight matrix with th e Hopfield network is proposed. The effectiveness of this algorithm is demo nstrated by computer simulation and optical implementation. In the optical implementation of the neural network the interconnection weights sire biase d to yield a nonnegative weight matrix. Moreover, a threshold subchannel is added so that the system can realize, in real time, the bipolar weighted s ummation in a single channel. Preliminary experimental results obtained fro m the applications in associative memories and, multitarget; classification with rotation invariance are shown. (C) 1999 Optical Society of America.