title: Categorical Perception and the Evolution of Supervised Learning in Neural Nets creator: Harnad, Stevan creator: Hanson, S.J. creator: Lubin, J. subject: Neural Nets subject: Perceptual Cognitive Psychology description: Some of the features of animal and human categorical perception (CP) for color, pitch and speech are exhibited by neural net simulations of CP with one-dimensional inputs: When a backprop net is trained to discriminate and then categorize a set of stimuli, the second task is accomplished by "warping" the similarity space (compressing within-category distances and expanding between-category distances). This natural side-effect also occurs in humans and animals. Such CP categories, consisting of named, bounded regions of similarity space, may be the ground level out of which higher-order categories are constructed; nets are one possible candidate for the mechanism that learns the sensorimotor invariants that connect arbitrary names (elementary symbols?) to the nonarbitrary shapes of objects. This paper examines how and why such compression/expansion effects occur in neural nets. publisher: Symposium on Symbol Grounding: Problems and Practice, Stanford University contributor: Powers, D. W. contributor: Reeker, L. date: 1991 type: Book Chapter type: NonPeerReviewed format: text/html identifier: http://cogprints.org/1579/1/harnad91.cpnets.html identifier: Harnad, Stevan and Hanson, S.J. and Lubin, J. (1991) Categorical Perception and the Evolution of Supervised Learning in Neural Nets. [Book Chapter] relation: http://cogprints.org/1579/