Return to search

Fast Learning by Bounding Likelihoods in Sigmoid Type Belief Networks

Sigmoid type belief networks, a class of probabilistic neural networks, provide a natural framework for compactly representing probabilistic information in a variety of unsupervised and supervised learning problems. Often the parameters used in these networks need to be learned from examples. Unfortunately, estimating the parameters via exact probabilistic calculations (i.e, the EM-algorithm) is intractable even for networks with fairly small numbers of hidden units. We propose to avoid the infeasibility of the E step by bounding likelihoods instead of computing them exactly. We introduce extended and complementary representations for these networks and show that the estimation of the network parameters can be made fast (reduced to quadratic optimization) by performing the estimation in either of the alternative domains. The complementary networks can be used for continuous density estimation as well.

Identiferoai:union.ndltd.org:MIT/oai:dspace.mit.edu:1721.1/7189
Date09 February 1996
CreatorsJaakkola, Tommi S., Saul, Lawrence K., Jordan, Michael I.
Source SetsM.I.T. Theses and Dissertation
Languageen_US
Detected LanguageEnglish
Format7 p., 197474 bytes, 292170 bytes, application/postscript, application/pdf
RelationAIM-1560, CBCL-129

Page generated in 0.0023 seconds