Fuzzy ARTMAP: A neural network architecture for incremental supervised learning of analog multidimensional maps

TLDR

The paper introduces a neural network architecture for incremental supervised learning of recognition categories and multidimensional maps from arbitrary analog or binary input sequences. The proposed fuzzy ARTMAP blends fuzzy logic with adaptive resonance theory, employing fuzzy subsethood computations for category choice, resonance, and learning, and is evaluated through simulations on geometric, function‑approximation, and letter‑recognition tasks, as well as comparisons with other algorithms. Simulations demonstrate that fuzzy ARTMAP performs competitively with benchmark backpropagation and generic algorithm systems across four task classes.

Abstract

A neural network architecture is introduced for incremental supervised learning of recognition categories and multidimensional maps in response to arbitrary sequences of analog or binary input vectors, which may represent fuzzy or crisp sets of features. The architecture, called fuzzy ARTMAP, achieves a synthesis of fuzzy logic and adaptive resonance theory (ART) neural networks by exploiting a close formal similarity between the computations of fuzzy subsethood and ART category choice, resonance, and learning. Four classes of simulation illustrated fuzzy ARTMAP performance in relation to benchmark backpropagation and generic algorithm systems. These simulations include finding points inside versus outside a circle, learning to tell two spirals apart, incremental approximation of a piecewise-continuous function, and a letter recognition database. The fuzzy ARTMAP system is also compared with Salzberg's NGE systems and with Simpson's FMMC system.