Journal article icon

Journal article

Hybrid variational/Gibbs collapsed inference in topic models

Abstract:

Variational Bayesian inference and (collapsed) Gibbs sampling are the two important classes of inference algorithms for Bayesian networks. Both have their advantages and disadvantages: collapsed Gibbs sampling is unbiased but is also inefficient for large count values and requires averaging over many samples to reduce variance. On the other hand, variational Bayesian inference is efficient and accurate for large count values but suffers from bias for small counts. We propose a hybrid algorith...

Expand abstract

Actions


Journal:
Proceedings of the 24th Conference on Uncertainty in Artificial Intelligence, UAI 2008 More from this journal
Pages:
587-594
Publication date:
2008-01-01
Language:
English
Pubs id:
pubs:353247
UUID:
uuid:8f5c433e-90ff-4c94-9006-1bd05d78531b
Local pid:
pubs:353247
Source identifiers:
353247
Deposit date:
2013-11-16

Terms of use


Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP