Training Products of Experts by Minimizing Contrastive

Training Products of Experts by Minimizing Contrastive (2002)

Cached

Download Links

[www-cse.ucsd.edu]

Other Repositories/Bibliography

DBLP

by Divergence Gcnu Tr , Geoffrey E. Hinton

Venue:

Neural Computation

Versions

Version History

Metadata Version 3

User correction supplied by SystemCorrections

Datum	Value	Source
TITLE	Training Products of Experts by Minimizing Contrastive	user correction - Legacy Corrections
AUTHOR NAME	Divergence Gcnu Tr	user correction - Legacy Corrections
AUTHOR NAME	Geoffrey E. Hinton	user correction - Legacy Corrections
AUTHOR AFFIL	Gatshy Computational Neuroscience Unit; University College London	user correction - Legacy Corrections
AUTHOR ADDR	17 Queen Square, London WC1N 3AR, U.K.	user correction - Legacy Corrections
ABSTRACT	It is possible to combine multiple probabilistic models of the same data by multiplying their probability distributions together and then renormalizing. This is a very efficient way to model high-dimensional data which simultaneously satisfies many different lowdimensional constraints because each individual expert model can focus on giving high probability to data vectors that satisfy just one of the constraints. Data vectors that satisfy this one constraint but violate other constraints will be ruled out by their low probability under the other experts. Training a product of experts appears difficult because, in addition to maximizing the probability that each individual expert assigns to the observed data, it is necessary to make the experts be as different as possible. This ensures that the product of their distributions is small which allows the renormalization to magnify the probability of the data under the product of experts model. Fortunately, if the individual experts are tractable there is an efficient way to train a product of experts.	user correction - Legacy Corrections
YEAR	2002	INFERENCE
VENUE	Neural Computation	INFERENCE
VENUE TYPE	JOURNAL	INFERENCE
PAGES	2002	INFERENCE
VOLUME	14	INFERENCE
TECH	Technical Report GCNU TR 2000-004	INFERENCE
CITATIONS	11 found	ParsCit 1.0