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.
@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},
address = {Long Beach, California, USA},
month = {09--15 Jun},
publisher = {PMLR},
pdf = {http://proceedings.mlr.press/v97/tavares19a/tavares19a.pdf},
url = {http://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.}
}
%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
%J Proceedings of Machine Learning Research
%P 6186--6195
%U http://proceedings.mlr.press
%V 97
%W PMLR
%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.
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 PMLR 97:6186-6195
This site last compiled Mon, 16 Sep 2019 16:05:04 +0000