Home > Research > Publications & Outputs > Fast variational inference in the conjugate exp...
View graph of relations

Fast variational inference in the conjugate exponential family

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Published

Standard

Fast variational inference in the conjugate exponential family. / Hensman, James; Rattray, Magnus; Lawrence, Neil D.
Advances in Neural Information Processing Systems. Vol. 4 2012. p. 2888-2896.

Research output: Contribution in Book/Report/Proceedings - With ISBN/ISSNConference contribution/Paperpeer-review

Harvard

Hensman, J, Rattray, M & Lawrence, ND 2012, Fast variational inference in the conjugate exponential family. in Advances in Neural Information Processing Systems. vol. 4, pp. 2888-2896, 26th Annual Conference on Neural Information Processing Systems 2012, NIPS 2012, Lake Tahoe, NV, United States, 3/12/12.

APA

Hensman, J., Rattray, M., & Lawrence, N. D. (2012). Fast variational inference in the conjugate exponential family. In Advances in Neural Information Processing Systems (Vol. 4, pp. 2888-2896)

Vancouver

Hensman J, Rattray M, Lawrence ND. Fast variational inference in the conjugate exponential family. In Advances in Neural Information Processing Systems. Vol. 4. 2012. p. 2888-2896

Author

Hensman, James ; Rattray, Magnus ; Lawrence, Neil D. / Fast variational inference in the conjugate exponential family. Advances in Neural Information Processing Systems. Vol. 4 2012. pp. 2888-2896

Bibtex

@inproceedings{308af4f505f248309b5186e80fa66670,
title = "Fast variational inference in the conjugate exponential family",
abstract = "We present a general method for deriving collapsed variational inference algorithms for probabilistic models in the conjugate exponential family. Our method unifies many existing approaches to collapsed variational inference. Our collapsed variational inference leads to a new lower bound on the marginal likelihood. We exploit the information geometry of the bound to derive much faster optimization methods based on conjugate gradients for these models. Our approach is very general and is easily applied to any model where the mean field update equations have been derived. Empirically we show significant speed-ups for probabilistic inference using our bound.",
author = "James Hensman and Magnus Rattray and Lawrence, {Neil D.}",
year = "2012",
language = "English",
isbn = "9781627480031",
volume = "4",
pages = "2888--2896",
booktitle = "Advances in Neural Information Processing Systems",
note = "26th Annual Conference on Neural Information Processing Systems 2012, NIPS 2012 ; Conference date: 03-12-2012 Through 06-12-2012",

}

RIS

TY - GEN

T1 - Fast variational inference in the conjugate exponential family

AU - Hensman, James

AU - Rattray, Magnus

AU - Lawrence, Neil D.

PY - 2012

Y1 - 2012

N2 - We present a general method for deriving collapsed variational inference algorithms for probabilistic models in the conjugate exponential family. Our method unifies many existing approaches to collapsed variational inference. Our collapsed variational inference leads to a new lower bound on the marginal likelihood. We exploit the information geometry of the bound to derive much faster optimization methods based on conjugate gradients for these models. Our approach is very general and is easily applied to any model where the mean field update equations have been derived. Empirically we show significant speed-ups for probabilistic inference using our bound.

AB - We present a general method for deriving collapsed variational inference algorithms for probabilistic models in the conjugate exponential family. Our method unifies many existing approaches to collapsed variational inference. Our collapsed variational inference leads to a new lower bound on the marginal likelihood. We exploit the information geometry of the bound to derive much faster optimization methods based on conjugate gradients for these models. Our approach is very general and is easily applied to any model where the mean field update equations have been derived. Empirically we show significant speed-ups for probabilistic inference using our bound.

M3 - Conference contribution/Paper

AN - SCOPUS:84877726166

SN - 9781627480031

VL - 4

SP - 2888

EP - 2896

BT - Advances in Neural Information Processing Systems

T2 - 26th Annual Conference on Neural Information Processing Systems 2012, NIPS 2012

Y2 - 3 December 2012 through 6 December 2012

ER -