Application of recurrent approximation to the synthesis of neural network for control of processes phototherapy

摘要

Three examples of problems of analytical learning as standardizated elements for design multy-layers neural network with any complicated configuration of input-output vector-function are consider. Solutions for synaptic weight coefficients (SWC) as recurrent sequence are represented in analytic form. Convergence of recurrent approximation for two schemes of approaches by a linear and quadratic curve are proposed and discussed for System Support Decision Making (SSDM) of physiotherapy with artificial neural networks (ANN).