A New Approach to Probabilistic Programming Inference

Frank Wood, Jan Willem Meent, Vikash Mansinghka
Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, PMLR 33:1024-1032, 2014.

Abstract

We introduce and demonstrate a new approach to inference in expressive probabilistic programming languages based on particle Markov chain Monte Carlo. Our approach is easy to implement and to parallelize, applies to Turing-complete probabilistic programming languages, and supports accurate inference in models that make use of complex control flow, including stochastic recursion, as well as primitives from nonparametric Bayesian statistics. Our experiments show that this approach can be more efficient than previously introduced single-site Metropolis-Hastings samplers.

Cite this Paper

BibTeX
@InProceedings{pmlr-v33-wood14, title = {{A New Approach to Probabilistic Programming Inference}}, author = {Wood, Frank and Meent, Jan Willem and Mansinghka, Vikash}, booktitle = {Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics}, pages = {1024--1032}, 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/wood14.pdf}, url = {https://proceedings.mlr.press/v33/wood14.html}, abstract = {We introduce and demonstrate a new approach to inference in expressive probabilistic programming languages based on particle Markov chain Monte Carlo. Our approach is easy to implement and to parallelize, applies to Turing-complete probabilistic programming languages, and supports accurate inference in models that make use of complex control flow, including stochastic recursion, as well as primitives from nonparametric Bayesian statistics. Our experiments show that this approach can be more efficient than previously introduced single-site Metropolis-Hastings samplers.} }
Endnote
%0 Conference Paper %T A New Approach to Probabilistic Programming Inference %A Frank Wood %A Jan Willem Meent %A Vikash Mansinghka %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-wood14 %I PMLR %P 1024--1032 %U https://proceedings.mlr.press/v33/wood14.html %V 33 %X We introduce and demonstrate a new approach to inference in expressive probabilistic programming languages based on particle Markov chain Monte Carlo. Our approach is easy to implement and to parallelize, applies to Turing-complete probabilistic programming languages, and supports accurate inference in models that make use of complex control flow, including stochastic recursion, as well as primitives from nonparametric Bayesian statistics. Our experiments show that this approach can be more efficient than previously introduced single-site Metropolis-Hastings samplers.
RIS
TY - CPAPER TI - A New Approach to Probabilistic Programming Inference AU - Frank Wood AU - Jan Willem Meent AU - Vikash Mansinghka 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-wood14 PB - PMLR DP - Proceedings of Machine Learning Research VL - 33 SP - 1024 EP - 1032 L1 - http://proceedings.mlr.press/v33/wood14.pdf UR - https://proceedings.mlr.press/v33/wood14.html AB - We introduce and demonstrate a new approach to inference in expressive probabilistic programming languages based on particle Markov chain Monte Carlo. Our approach is easy to implement and to parallelize, applies to Turing-complete probabilistic programming languages, and supports accurate inference in models that make use of complex control flow, including stochastic recursion, as well as primitives from nonparametric Bayesian statistics. Our experiments show that this approach can be more efficient than previously introduced single-site Metropolis-Hastings samplers. ER -
APA
Wood, F., Meent, J.W. & Mansinghka, V.. (2014). A New Approach to Probabilistic Programming Inference. Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 33:1024-1032 Available from https://proceedings.mlr.press/v33/wood14.html.

Related Material