Fuzzy ART: an adaptive resonance algorithm for rapid, stable classification of analog patterns

• Published in: IJCNN-91-Seattle International Joint Conference on Neural Networks Vol. ii; pp. 411 - 416 vol.2
• Main Authors: Carpenter, G.A., Grossberg, S., Rosen, D.B.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 1991
• Subjects: Adaptive systems; Analog computers; Data preprocessing; Equations; Fuzzy set theory; Fuzzy sets; Fuzzy systems; Resonance; Subspace constraints
• ISBN: 0780301641; 9780780301641
• DOI: 10.1109/IJCNN.1991.155368

A fuzzy ART (adaptive resonance theory) system is introduced which incorporates computations from fuzzy set theory into ART 1. For example, the intersection ( intersection ) operator used in ART 1 learning is replaced by the MIN operator ( V-product ) of fuzzy set theory. Fuzzy ART reduces to ART 1 in response to binary input vectors, but can also learn stable categories in response to analog input vectors. In particular, the MIN operator reduces to the intersection operator in the binary case. Learning is stable because all adaptive weights can only decrease in time. A preprocessing step, called complement coding, uses on-cell and off-cell responses to prevent category proliferation. Complement coding normalizes input vectors while preserving the amplitude of individual feature activations.< >