Predicate Exchange: Inference with Declarative Knowledge

Zenna Tavares, Javier Burroni, Edgar Minasyan, Armando Solar-Lezama, Rajesh Ranganath
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:6186-6195, 2019.

Abstract

Programming languages allow us to express complex predicates, but existing inference methods are unable to condition probabilistic models on most of them. To support a broader class of predicates, we develop an inference procedure called predicate exchange, which softens predicates. A soft predicate quantifies the extent to which values of model variables are consistent with its hard counterpart. We substitute the likelihood term in the Bayesian posterior with a soft predicate, and develop a variant of replica exchange MCMC to draw posterior samples. We implement predicate exchange as a language agnostic tool which performs a nonstandard execution of a probabilistic program. We demonstrate the approach on sequence models of health and inverse rendering.

Cite this Paper

BibTeX
@InProceedings{pmlr-v97-tavares19a, title = {Predicate Exchange: Inference with Declarative Knowledge}, author = {Tavares, Zenna and Burroni, Javier and Minasyan, Edgar and Solar-Lezama, Armando and Ranganath, Rajesh}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {6186--6195}, year = {2019}, editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan}, volume = {97}, series = {Proceedings of Machine Learning Research}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/tavares19a/tavares19a.pdf}, url = {https://proceedings.mlr.press/v97/tavares19a.html}, abstract = {Programming languages allow us to express complex predicates, but existing inference methods are unable to condition probabilistic models on most of them. To support a broader class of predicates, we develop an inference procedure called predicate exchange, which softens predicates. A soft predicate quantifies the extent to which values of model variables are consistent with its hard counterpart. We substitute the likelihood term in the Bayesian posterior with a soft predicate, and develop a variant of replica exchange MCMC to draw posterior samples. We implement predicate exchange as a language agnostic tool which performs a nonstandard execution of a probabilistic program. We demonstrate the approach on sequence models of health and inverse rendering.} }
Endnote
%0 Conference Paper %T Predicate Exchange: Inference with Declarative Knowledge %A Zenna Tavares %A Javier Burroni %A Edgar Minasyan %A Armando Solar-Lezama %A Rajesh Ranganath %B Proceedings of the 36th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Ruslan Salakhutdinov %F pmlr-v97-tavares19a %I PMLR %P 6186--6195 %U https://proceedings.mlr.press/v97/tavares19a.html %V 97 %X Programming languages allow us to express complex predicates, but existing inference methods are unable to condition probabilistic models on most of them. To support a broader class of predicates, we develop an inference procedure called predicate exchange, which softens predicates. A soft predicate quantifies the extent to which values of model variables are consistent with its hard counterpart. We substitute the likelihood term in the Bayesian posterior with a soft predicate, and develop a variant of replica exchange MCMC to draw posterior samples. We implement predicate exchange as a language agnostic tool which performs a nonstandard execution of a probabilistic program. We demonstrate the approach on sequence models of health and inverse rendering.
APA
Tavares, Z., Burroni, J., Minasyan, E., Solar-Lezama, A. & Ranganath, R.. (2019). Predicate Exchange: Inference with Declarative Knowledge. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:6186-6195 Available from https://proceedings.mlr.press/v97/tavares19a.html.

Related Material