Access this book
Tax calculation will be finalised at checkout
Other ways to access
About this book
Artificial "neural networks" are widely used as flexible models for classification and regression applications, but questions remain about how the power of these models can be safely exploited when training data is limited. This book demonstrates how Bayesian methods allow complex neural network models to be used without fear of the "overfitting" that can occur with traditional training methods. Insight into the nature of these complex Bayesian models is provided by a theoretical investigation of the priors over functions that underlie them. A practical implementation of Bayesian neural network learning using Markov chain Monte Carlo methods is also described, and software for it is freely available over the Internet. Presupposing only basic knowledge of probability and statistics, this book should be of interest to researchers in statistics, engineering, and artificial intelligence.
Similar content being viewed by others
Keywords
Table of contents (5 chapters)
Authors and Affiliations
Bibliographic Information
Book Title: Bayesian Learning for Neural Networks
Authors: Radford M. Neal
Series Title: Lecture Notes in Statistics
DOI: https://doi.org/10.1007/978-1-4612-0745-0
Publisher: Springer New York, NY
-
eBook Packages: Springer Book Archive
Copyright Information: Springer Science+Business Media New York 1996
Softcover ISBN: 978-0-387-94724-2Published: 09 August 1996
eBook ISBN: 978-1-4612-0745-0Published: 06 December 2012
Series ISSN: 0930-0325
Series E-ISSN: 2197-7186
Edition Number: 1
Number of Pages: 204
Topics: Probability Theory and Stochastic Processes, Statistics for Engineering, Physics, Computer Science, Chemistry and Earth Sciences, Artificial Intelligence, Simulation and Modeling