A Level-set Hit-and-run Sampler for Quasi-Concave Distributions

Shane Jensen, Dean Foster
Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, PMLR 33:439-447, 2014.

Abstract

We develop a new sampling strategy that uses the hit-and-run algorithm within level sets of a target density. Our method can be applied to any quasi-concave density, which covers a broad class of models. Standard sampling methods often perform poorly on densities that are high-dimensional or multi-modal. Our level set sampler performs well in high-dimensional settings, which we illustrate on a spike-and-slab mixture model. We also extend our method to exponentially-tilted quasi-concave densities, which arise in Bayesian models consisting of a log-concave likelihood and quasi-concave prior density. We illustrate our exponentially-tilted level-set sampler on a Cauchy-normal model where our sampler is better able to handle a high-dimensional and multi-modal posterior distribution compared to Gibbs sampling and Hamiltonian Monte Carlo.

Cite this Paper


BibTeX
@InProceedings{pmlr-v33-jensen14, title = {{A Level-set Hit-and-run Sampler for Quasi-Concave Distributions}}, author = {Jensen, Shane and Foster, Dean}, booktitle = {Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics}, pages = {439--447}, year = {2014}, editor = {Kaski, Samuel and Corander, Jukka}, volume = {33}, series = {Proceedings of Machine Learning Research}, address = {Reykjavik, Iceland}, month = {22--25 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v33/jensen14.pdf}, url = {https://proceedings.mlr.press/v33/jensen14.html}, abstract = {We develop a new sampling strategy that uses the hit-and-run algorithm within level sets of a target density. Our method can be applied to any quasi-concave density, which covers a broad class of models. Standard sampling methods often perform poorly on densities that are high-dimensional or multi-modal. Our level set sampler performs well in high-dimensional settings, which we illustrate on a spike-and-slab mixture model. We also extend our method to exponentially-tilted quasi-concave densities, which arise in Bayesian models consisting of a log-concave likelihood and quasi-concave prior density. We illustrate our exponentially-tilted level-set sampler on a Cauchy-normal model where our sampler is better able to handle a high-dimensional and multi-modal posterior distribution compared to Gibbs sampling and Hamiltonian Monte Carlo.} }
Endnote
%0 Conference Paper %T A Level-set Hit-and-run Sampler for Quasi-Concave Distributions %A Shane Jensen %A Dean Foster %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-jensen14 %I PMLR %P 439--447 %U https://proceedings.mlr.press/v33/jensen14.html %V 33 %X We develop a new sampling strategy that uses the hit-and-run algorithm within level sets of a target density. Our method can be applied to any quasi-concave density, which covers a broad class of models. Standard sampling methods often perform poorly on densities that are high-dimensional or multi-modal. Our level set sampler performs well in high-dimensional settings, which we illustrate on a spike-and-slab mixture model. We also extend our method to exponentially-tilted quasi-concave densities, which arise in Bayesian models consisting of a log-concave likelihood and quasi-concave prior density. We illustrate our exponentially-tilted level-set sampler on a Cauchy-normal model where our sampler is better able to handle a high-dimensional and multi-modal posterior distribution compared to Gibbs sampling and Hamiltonian Monte Carlo.
RIS
TY - CPAPER TI - A Level-set Hit-and-run Sampler for Quasi-Concave Distributions AU - Shane Jensen AU - Dean Foster BT - Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics DA - 2014/04/02 ED - Samuel Kaski ED - Jukka Corander ID - pmlr-v33-jensen14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 33 SP - 439 EP - 447 L1 - http://proceedings.mlr.press/v33/jensen14.pdf UR - https://proceedings.mlr.press/v33/jensen14.html AB - We develop a new sampling strategy that uses the hit-and-run algorithm within level sets of a target density. Our method can be applied to any quasi-concave density, which covers a broad class of models. Standard sampling methods often perform poorly on densities that are high-dimensional or multi-modal. Our level set sampler performs well in high-dimensional settings, which we illustrate on a spike-and-slab mixture model. We also extend our method to exponentially-tilted quasi-concave densities, which arise in Bayesian models consisting of a log-concave likelihood and quasi-concave prior density. We illustrate our exponentially-tilted level-set sampler on a Cauchy-normal model where our sampler is better able to handle a high-dimensional and multi-modal posterior distribution compared to Gibbs sampling and Hamiltonian Monte Carlo. ER -
APA
Jensen, S. & Foster, D.. (2014). A Level-set Hit-and-run Sampler for Quasi-Concave Distributions. Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 33:439-447 Available from https://proceedings.mlr.press/v33/jensen14.html.

Related Material