Radial Basis Funct ion Art ificia l Neural Netw orks

An artificial neural network (ANN) is an information-processing paradigm that is designed to emulate some of the observed properties of the mammalian brain. First proposed in the early 1950's it was not until the technology revolution of the 1980's that a multitude of alternative ANN methods were spawned to solve a wide variety of complex real-world problems. Today, the contemporary literature on ANNs is replete with successful reports of applications to problems that are too complex for conventional algorithmic methods or for which an algorithmic specification is too complex for practical implementation. The robustness of the ANN method under difficult modeling conditions is also well documented. For example, ANNs have proven extremely resilient against distortions introduced by noisy data. In short, the ANN paradigm has a developed a track-record as a good pattern recognition engine, a robust classifier, and an expert functional agent in prediction and system modeling where the physical processes are not easily understood or are highly complex.