Supervised learning of probability distributions by neural networksSupervised learning algorithms for feedforward neural networks are investigated analytically. The back-propagation algorithm described by Werbos (1974), Parker (1985), and Rumelhart et al. (1986) is generalized by redefining the values of the input and output neurons as probabilities. The synaptic weights are then varied to follow gradients in the logarithm of likelihood rather than in the error. This modification is shown to provide a more rigorous theoretical basis for the algorithm and to permit more accurate predictions. A typical application involving a medical-diagnosis expert system is discussed.
Document ID
19890041637
Acquisition Source
Legacy CDMS
Document Type
Conference Paper
Authors
Baum, Eric B. (California Institute of Technology Jet Propulsion Laboratory, Pasadena, United States)
Wilczek, Frank (Harvard University Cambridge, MA, United States)
Date Acquired
August 14, 2013
Publication Date
January 1, 1988
Subject Category
Cybernetics
Meeting Information
Meeting: IEEE Conference on Neural Information Processing Systems