RBF fuzzy-artmap: a new fuzzy neural network for robust on-line learning and identification of patterns
Gérson Tontini, A.A. De Queiroz · 2002
One of the most important difficulties in pattern identification with neural networks is the sensibility to the presentation order of the training patterns. This paper presents a new network, RBF fuzzy-artmap, capable of online incremental learning, not sensible to the presentation order of training patterns. It is a hybrid network, formed by a RBF network and a fuzzy-artmap network Because it uses the learning method of the RBF network in a fuzzy-artmap structure, it overcomes limitations of both, RBF and fuzzy-artmap networks. The RBF fuzzy-artmap network is 98% less sensible to the presentation order of training patterns than the fuzzy-artmap network.