Tilted Variational Bayes

James Hensman, Max Zwiessele, Neil Lawrence
; Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, PMLR 33:356-364, 2014.

Abstract

We present a novel method for approximate inference. Using some of the constructs from expectation propagation (EP), we derive a lower bound of the marginal likelihood in a similar fashion to variational Bayes (VB). The method combines some of the benefits of VB and EP: it can be used with light-tailed likelihoods (where traditional VB fails), and it provides a lower bound on the marginal likelihood. We apply the method to Gaussian process classification, a situation where the Kullback-Leibler divergence minimized in traditional VB can be infinite, and to robust Gaussian process regression, where the inference process is dramatically simplified in comparison to EP. Code to reproduce all the experiments can be found at github.com/SheffieldML/TVB.

Cite this Paper


BibTeX
@InProceedings{pmlr-v33-hensman14, title = {{Tilted Variational Bayes}}, author = {James Hensman and Max Zwiessele and Neil Lawrence}, booktitle = {Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics}, pages = {356--364}, year = {2014}, editor = {Samuel Kaski and Jukka Corander}, volume = {33}, series = {Proceedings of Machine Learning Research}, address = {Reykjavik, Iceland}, month = {22--25 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v33/hensman14.pdf}, url = {http://proceedings.mlr.press/v33/hensman14.html}, abstract = {We present a novel method for approximate inference. Using some of the constructs from expectation propagation (EP), we derive a lower bound of the marginal likelihood in a similar fashion to variational Bayes (VB). The method combines some of the benefits of VB and EP: it can be used with light-tailed likelihoods (where traditional VB fails), and it provides a lower bound on the marginal likelihood. We apply the method to Gaussian process classification, a situation where the Kullback-Leibler divergence minimized in traditional VB can be infinite, and to robust Gaussian process regression, where the inference process is dramatically simplified in comparison to EP. Code to reproduce all the experiments can be found at github.com/SheffieldML/TVB.} }
Endnote
%0 Conference Paper %T Tilted Variational Bayes %A James Hensman %A Max Zwiessele %A Neil Lawrence %B Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2014 %E Samuel Kaski %E Jukka Corander %F pmlr-v33-hensman14 %I PMLR %J Proceedings of Machine Learning Research %P 356--364 %U http://proceedings.mlr.press %V 33 %W PMLR %X We present a novel method for approximate inference. Using some of the constructs from expectation propagation (EP), we derive a lower bound of the marginal likelihood in a similar fashion to variational Bayes (VB). The method combines some of the benefits of VB and EP: it can be used with light-tailed likelihoods (where traditional VB fails), and it provides a lower bound on the marginal likelihood. We apply the method to Gaussian process classification, a situation where the Kullback-Leibler divergence minimized in traditional VB can be infinite, and to robust Gaussian process regression, where the inference process is dramatically simplified in comparison to EP. Code to reproduce all the experiments can be found at github.com/SheffieldML/TVB.
RIS
TY - CPAPER TI - Tilted Variational Bayes AU - James Hensman AU - Max Zwiessele AU - Neil Lawrence BT - Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics PY - 2014/04/02 DA - 2014/04/02 ED - Samuel Kaski ED - Jukka Corander ID - pmlr-v33-hensman14 PB - PMLR SP - 356 DP - PMLR EP - 364 L1 - http://proceedings.mlr.press/v33/hensman14.pdf UR - http://proceedings.mlr.press/v33/hensman14.html AB - We present a novel method for approximate inference. Using some of the constructs from expectation propagation (EP), we derive a lower bound of the marginal likelihood in a similar fashion to variational Bayes (VB). The method combines some of the benefits of VB and EP: it can be used with light-tailed likelihoods (where traditional VB fails), and it provides a lower bound on the marginal likelihood. We apply the method to Gaussian process classification, a situation where the Kullback-Leibler divergence minimized in traditional VB can be infinite, and to robust Gaussian process regression, where the inference process is dramatically simplified in comparison to EP. Code to reproduce all the experiments can be found at github.com/SheffieldML/TVB. ER -
APA
Hensman, J., Zwiessele, M. & Lawrence, N.. (2014). Tilted Variational Bayes. Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, in PMLR 33:356-364

Related Material