Probable networks and plausible predictions — a review of practical Bayesian methods for supervised neural networks
- 1 August 1995
- journal article
- review article
- Published by Informa UK Limited in Network: Computation in Neural Systems
- Vol. 6 (3), 469-505
- https://doi.org/10.1088/0954-898x/6/3/011
Abstract
Bayesian probability theory provides a unifying framework for data modelling. In this framework the overall aims are to find models that are well-matched to the data, and to use these models to make optimal predictions. Neural network learning is interpreted as an inference of the most probable parameters for the model, given the training data. The search in model space (i.e., the space of architectures, noise models, preprocessings, regularizers and weight decay constants) can then also be treated as an inference problem, in which we infer the relative probability of alternative models, given the data. This review describes practical techniques based on Gaussian approximations for implementation of these powerful methods for controlling, comparing and using adaptive networks.This publication has 14 references indexed in Scilit:
- Fast Exact Multiplication by the HessianNeural Computation, 1994
- Exact Calculation of the Hessian Matrix for the Multilayer PerceptronNeural Computation, 1992
- Probabilistic DisplaysPublished by Springer Science and Business Media LLC ,1991
- Sequential updating of conditional probabilities on directed graphical structuresNetworks, 1990
- From Laplace to Supernova SN 1987A: Bayesian Inference in AstrophysicsPublished by Springer Science and Business Media LLC ,1990
- The Vapnik-Chervonenkis Dimension: Information versus Complexity in LearningNeural Computation, 1989
- Arithmetic coding for data compressionCommunications of the ACM, 1987
- Learning representations by back-propagating errorsNature, 1986
- An Information Measure for ClassificationThe Computer Journal, 1968
- Probability, Frequency and Reasonable ExpectationAmerican Journal of Physics, 1946