Presentation 1997/2/7
A hierachical Bayesian approach of ARD(Automatic Relevance Determination) with M.L.P.
Y. Nakajima, T. Matsumoto,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Given data set D={t^m,x^m}^N_M=1 for training, there's a possibility that some of the input variables x^m are actually irrelevant to the prediction of the output variable. Weights are devided into groups which are connected to each input and a regularizer is assigned to each group. A hierachi Bayes approach is taken to perform ARD and regularizer comparison.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Neural Network / Hierachical Bayesian learning / ARD
Paper # NLP96-149,NC96-103
Date of Issue

Conference Information
Committee NC
Conference Date 1997/2/7(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A hierachical Bayesian approach of ARD(Automatic Relevance Determination) with M.L.P.
Sub Title (in English)
Keyword(1) Neural Network
Keyword(2) Hierachical Bayesian learning
Keyword(3) ARD
1st Author's Name Y. Nakajima
1st Author's Affiliation Department of Electrical, Electronics, and Computer Engeneering, Waseda University()
2nd Author's Name T. Matsumoto
2nd Author's Affiliation Department of Electrical, Electronics, and Computer Engeneering, Waseda University
Date 1997/2/7
Paper # NLP96-149,NC96-103
Volume (vol) vol.96
Number (no) 512
Page pp.pp.-
#Pages 8
Date of Issue