The Structure of Neural Network Posteriors

Neil D. Lawrence, Mehdi Azzouzi
, 2001.

Abstract

Exact inference in Bayesian neural networks is non analytic to compute and as a result approximate approaches such as the evidence procedure, Monte-Carlo sampling and variational inference have been proposed. In this paper we explore the structure of the posterior distributions in such a model through a new approximating distribution based on *mixtures* of Gaussian distributions and show how it may be implemented.

Cite this Paper

BibTeX
@Misc{Lawrence:structure01, title = {The Structure of Neural Network Posteriors}, author = {Lawrence, Neil D. and Azzouzi, Mehdi}, year = {2001}, pdf = {http://www.thelawrences.net/neil/mixture.pdf}, url = {http://inverseprobability.com/publications/lawrence-structure01.html}, abstract = {Exact inference in Bayesian neural networks is non analytic to compute and as a result approximate approaches such as the evidence procedure, Monte-Carlo sampling and variational inference have been proposed. In this paper we explore the structure of the posterior distributions in such a model through a new approximating distribution based on *mixtures* of Gaussian distributions and show how it may be implemented.}, note = {} }
Endnote
%0 Generic %T The Structure of Neural Network Posteriors %A Neil D. Lawrence %A Mehdi Azzouzi %D 2001 %F Lawrence:structure01 %U http://inverseprobability.com/publications/lawrence-structure01.html %X Exact inference in Bayesian neural networks is non analytic to compute and as a result approximate approaches such as the evidence procedure, Monte-Carlo sampling and variational inference have been proposed. In this paper we explore the structure of the posterior distributions in such a model through a new approximating distribution based on *mixtures* of Gaussian distributions and show how it may be implemented. %Z
RIS
TY - GEN TI - The Structure of Neural Network Posteriors AU - Neil D. Lawrence AU - Mehdi Azzouzi DA - 2001/01/01 ID - Lawrence:structure01 L1 - http://www.thelawrences.net/neil/mixture.pdf UR - http://inverseprobability.com/publications/lawrence-structure01.html AB - Exact inference in Bayesian neural networks is non analytic to compute and as a result approximate approaches such as the evidence procedure, Monte-Carlo sampling and variational inference have been proposed. In this paper we explore the structure of the posterior distributions in such a model through a new approximating distribution based on *mixtures* of Gaussian distributions and show how it may be implemented. N1 - ER -
APA
Lawrence, N.D. & Azzouzi, M.. (2001). The Structure of Neural Network Posteriors. Available from http://inverseprobability.com/publications/lawrence-structure01.html.

Related Material