Tractable Uncertainty for Structure Learning

Benjie Wang, Matthew R Wicker, Marta Kwiatkowska
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:23131-23150, 2022.

Abstract

Bayesian structure learning allows one to capture uncertainty over the causal directed acyclic graph (DAG) responsible for generating given data. In this work, we present Tractable Uncertainty for STructure learning (TRUST), a framework for approximate posterior inference that relies on probabilistic circuits as a representation of our posterior belief. In contrast to sample-based posterior approximations, our representation can capture a much richer space of DAGs, while being able to tractably answer a range of useful inference queries. We empirically demonstrate how probabilistic circuits can be used to as an augmented representation for structure learning methods, leading to improvement in both the quality of inferred structures and posterior uncertainty. Experimental results also demonstrate the improved representational capacity of TRUST, outperforming competing methods on conditional query answering.

Cite this Paper


BibTeX
@InProceedings{pmlr-v162-wang22ad, title = {Tractable Uncertainty for Structure Learning}, author = {Wang, Benjie and Wicker, Matthew R and Kwiatkowska, Marta}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {23131--23150}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/wang22ad/wang22ad.pdf}, url = {https://proceedings.mlr.press/v162/wang22ad.html}, abstract = {Bayesian structure learning allows one to capture uncertainty over the causal directed acyclic graph (DAG) responsible for generating given data. In this work, we present Tractable Uncertainty for STructure learning (TRUST), a framework for approximate posterior inference that relies on probabilistic circuits as a representation of our posterior belief. In contrast to sample-based posterior approximations, our representation can capture a much richer space of DAGs, while being able to tractably answer a range of useful inference queries. We empirically demonstrate how probabilistic circuits can be used to as an augmented representation for structure learning methods, leading to improvement in both the quality of inferred structures and posterior uncertainty. Experimental results also demonstrate the improved representational capacity of TRUST, outperforming competing methods on conditional query answering.} }
Endnote
%0 Conference Paper %T Tractable Uncertainty for Structure Learning %A Benjie Wang %A Matthew R Wicker %A Marta Kwiatkowska %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-wang22ad %I PMLR %P 23131--23150 %U https://proceedings.mlr.press/v162/wang22ad.html %V 162 %X Bayesian structure learning allows one to capture uncertainty over the causal directed acyclic graph (DAG) responsible for generating given data. In this work, we present Tractable Uncertainty for STructure learning (TRUST), a framework for approximate posterior inference that relies on probabilistic circuits as a representation of our posterior belief. In contrast to sample-based posterior approximations, our representation can capture a much richer space of DAGs, while being able to tractably answer a range of useful inference queries. We empirically demonstrate how probabilistic circuits can be used to as an augmented representation for structure learning methods, leading to improvement in both the quality of inferred structures and posterior uncertainty. Experimental results also demonstrate the improved representational capacity of TRUST, outperforming competing methods on conditional query answering.
APA
Wang, B., Wicker, M.R. & Kwiatkowska, M.. (2022). Tractable Uncertainty for Structure Learning. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:23131-23150 Available from https://proceedings.mlr.press/v162/wang22ad.html.

Related Material