Adaptive learning of polynomial networks: genetic programming, backpropagation and Bayesian methods (Genetic and Evolutionary Computation Series)

摘要

The polynomial neural network (PNN) framework described in this book has its origins in Ivakhnenko's "group method of data handling," or GMDH, from the early 1970s. The GMDH technique and its variants, based upon a neural network-like hierarchy of simple multivariate polynomial activation functions, have been applied to a wide variety of problems in forecasting, approximation, clustering, pattern recognition, and other areas.