Browse Bar: Browse by Author | Browse by Category | Browse by Citation | Advanced Search
Author(s): Carpenter, G.A. | Grossberg, S. |
Year: 1990
Citation: In M.L. Commons, R.J. Herrnstein, S.M. Kosslyn, & D.B. Mumford (Eds.), Quantitative Analyses of Behavior, IX: Computational and Clinical Approaches to Pattern Recognition and Concept Formation, Hillsdale, NJ: Erlbaum Associates, 23-49.
Abstract: This chapter describes how cognitive recognition codes can be learned in response to a temporal stream of input patterns. This self-organizing learning process automatically buffers, or self-stabilizes, its learning against recoding by the blooming buzzing confusion of irrelevant experience, no matter how many input patterns are processed, and no matter how complex these input patterns may be, until it utilizes its full memory capacity. These results are part of an extensive development of Grossberg?s adaptive resonance theory, or ART (Carpenter & Grossberg, 1987 a,b,c, 1988). The ART model discussed herein is called ART 1 (Carpenter & Grossberg, 1987a, 1988). It was developed to explain how a world of binary input patterns can be learned and recognized. A more recent model, called ART 2 was developed to deal with an analog or binary input world (Carpenter & Grossberg, 1987 b,c). An account of the historical factors leading to the introduction of adaptive resonance theory and its relationship to other neural network models, such as competitive learning, interactive activation, and back propagation, is found in Grossberg (1987a).
Topics:
Machine Learning,
Models:
ART 1,
ART 2 / Fuzzy ART,
