Abstract
Difficulties of the Hopfield-Tank (H-T) model reported by Wilson-Pawley [Bio-Cybern. 58, 63 (1988)] in a scale up Travelling Salesman Problem (TSP) have been clarified and repudiated. H-T [Bio-Cybern. 52, 141 (1985)] have successfully integrated the decaying Hamiltonian, H(Vi; a, b, c) equation dUi/dt =(Ui/t) – [∂H(Vj; a, b, c)/∂Vi] for the neuron input Ui given the neuron output Vi by a sigmoid mapping Vi = g(Ui).An improvement is the conservative dynamics eliminating time-consuming computation for every iteration of each analog neuron (good for a buzz-booming real world) and the self-denying interconnect Tii < 0 (necessary for VLSI resistive coupling). H-T dynamics is weakened by degenerate ground states due to those deliberately imprecise and inhibitory constraints (modeling the Mother Nature). In artificial problems, a hybrid neuron model consisting of a binary neuron with analog interconnects inputting intercity distances may be efficient. A decisive yes-visit-or-no-visit binary neuron may be better than a wishy-washy sigmoid neuron, for it can revive the ball-balancing game rolling over the energy landscape. The zero diagonal interconnect matrix means no self-talk (traceless Tii = 0), which is best for the absolute convergence theorem (self-promoting Tii > 0 may be OK, but self-denying Tii < 0 is bad) for asynchronous and democratic voting updates by peer neurons excluding itself.
© 1988 Optical Society of America
PDF ArticleMore Like This
L. M. Kani and J. C. Dainty
FJ4 OSA Annual Meeting (FIO) 1988
Inbarasan Muniraj, Changliang Guo, Ra’ed Malallah, Derek Cassidy, Liang Zhao, James P Ryle, John J Healy, and John T Sheridan
W1A.4 Digital Holography and Three-Dimensional Imaging (DH) 2017
T. L. Jong, Stephen G. Batsell, John F. Walkup, and Thomas F. Krile
THHH3 OSA Annual Meeting (FIO) 1988