Proceedings of the Annual Meeting of the Cognitive Science Society

A general learning rule, “BCM-δ ”, is proposed that subsumes both unsupervised learning as a form of the BCMrule (Bienenstock, Cooper, Munro, 1982; Munro, 1984) and the delta rule (Rosenblatt, 1958; Rumelhart, Hinton, and Williams,1986). The “BCM-δ ” unit is composed of two subunits, T and L, each integrating distinct input streams across distinct setsof synapses. The two subunits follow a common Hebb-like learning procedure that reduces to an unsupervised rule for the Tsubunit and a supervised rule for the L subunit in which the T response is the training signal. This model suggests a neurallyplausible mechanism for the shaping of concepts by labels. More generally, stimuli from one modality can shape the responseproperties of a unit to another modality using a framework that is biologically plausible and gives clues to the source of ateaching signal for supervised learning.