Use of Feynman diagrams in large-scale neural networks with nonlinear behavior
José A. Martín-Pereda, Ana Pilar Gonzalez-Marcos · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1993
A new proposal to the study of large-scale neural networks is reported. It is based on the use of similar graphs to the Feynman diagrams. A first general theory is presented and some interpretations are given. A propagator, based on the Green's function of the neuron, is the basis of the method. Application to a simple case is reported.