Evolving Fuzzy Neural Networks - Algorithms, Applications and Biological Motivation

摘要

In the paper, the ECOS (Evolving Connectionist Systems) framework is used to develop a particular type of evolving neural networks - evolving fuzzy neural networks - EFuNNs. They can be trained in an on-line, incremental mode and are several order of magnitude faster than the ordinary neural networks trained with the backpropagation algorithm. This is illustrated on the task of adaptive phoneme recognition.