Learning from noisy information in FasArt and FasBack neuro-fuzzy systems

Neuro-fuzzy systems have been in the focus of recent research as a solution to jointly exploit the main features of fuzzy logic systems and neural net works. Within the application literature, neuro-fuzzy systems can be found as methods for function identification. This approach is supported by theor ems that guarantee the possibility of representing arbitrary functions by f uzzy systems. However, due to the fact that real data are often noisy, gene ration of accurate identifiers is presented as an important problem. Within the Adaptive Resonance Theory (ART), PROBART architecture has been propose d as a solution to this problem. After a detailed comparison of these archi tectures based on their design principles, the FasArt and FasBack models ar e proposed. They are neuro-fuzzy identifiers that offer a dual interpretati on, as fuzzy logic systems or neural networks. FasArt and FasBack can be tr ained on noisy data without need of change in their structure or data prepr ocessing. In the simulation work, a comparative study is carried out on the performances of Fuzzy ARTMAP, PROBART. FasArt and FasBack, focusing on pre diction error and network complexity. Results show that FasArt and FasBack clearly enhance the performance of other models in this important problem. (C) 2001 Elsevier Science Ltd. All rights reserved.