Mean Field Approximation in Bayesian Variable Selection
Disordered Systems and Neural Networks
2007-05-23 v1 Statistical Mechanics
Abstract
Variable selection for a multiple regression model (Noisy Linear Perceptron) is studied with a mean field approximation. In our Bayesian framework, variable selection is formulated as estimation of discrete parameters that indicate a subset of the explanatory variables. Then, a mean field approximation is introduced for the calculation of the posterior averages over the discrete parameters. An application to a real world example, Boston housing data, is shown.
Keywords
Cite
@article{arxiv.cond-mat/9808071,
title = {Mean Field Approximation in Bayesian Variable Selection},
author = {Yukito Iba},
journal= {arXiv preprint arXiv:cond-mat/9808071},
year = {2007}
}
Comments
4 pages, 2 figures(5 ps files), uses epsf.sty, iconip98.sty, to appear in the proceedings of ICONIP'98-Kitakyushu