Optical Realizations of Adaptive Resonance Theory Networks
Cornelia Denz · 1998
Since its conception in 1987 by G. Carpenter and S. Grossberg [15, 9], the adaptive resonance theory (ART) has become more and more attractive in solving a growing number of problems, especially in group-technology problems and in motion control. This is mainly due to its stable unsupervised learning properties and its ability to process large input-pattern fields. Successful electronic applications of ART used input fields in excess of 10 7 nodes.It seems that ART’s scaling properties are limited only by hardware and software implementations — a significant motivation for many researchers in the past years to exploit the features of ART in many different applications using optical realizations with their inherent potential of parallel processing. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.