A New Approach to Probabilistic Programming Inference

Frank D. Wood, Jan-Willem van de Meent, Vikash Mansinghka

AISTATS 2014 pp. 1024-1032

doi:10.14288/1.0044249 /aistats/2014/wood2014aistats-new/

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 simple to implement and easy to parallelize. It applies to Turing-complete probabilistic programming languages and supports accurate inference in models that make use of complex control ow, including stochastic recursion. It also includes primitives from Bayesian nonparametric statistics. Our experiments show that this approach can be more ecient than previously introduced single-site Metropolis-Hastings methods.

PDF AISTATS Semantic Scholar

Cite

Text

Wood et al. "A New Approach to Probabilistic Programming Inference." International Conference on Artificial Intelligence and Statistics, 2014. doi:10.14288/1.0044249

Markdown

[Wood et al. "A New Approach to Probabilistic Programming Inference." International Conference on Artificial Intelligence and Statistics, 2014.](https://mlanthology.org/aistats/2014/wood2014aistats-new/) doi:10.14288/1.0044249

BibTeX

@inproceedings{wood2014aistats-new,
  title     = {{A New Approach to Probabilistic Programming Inference}},
  author    = {Wood, Frank D. and van de Meent, Jan-Willem and Mansinghka, Vikash},
  booktitle = {International Conference on Artificial Intelligence and Statistics},
  year      = {2014},
  pages     = {1024-1032},
  doi       = {10.14288/1.0044249},
  url       = {https://mlanthology.org/aistats/2014/wood2014aistats-new/}
}