Tractable Uncertainty for Structure Learning
This work addresses uncertainty quantification in causal discovery, which is important for researchers and practitioners in fields like machine learning and statistics, though it appears incremental as it builds on existing structure learning methods with a new representation.
The paper tackles the problem of capturing uncertainty in Bayesian structure learning for causal directed acyclic graphs (DAGs) by introducing TRUST, a framework that uses probabilistic circuits for approximate posterior inference, leading to improvements in the quality of inferred structures and posterior uncertainty.
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 the representation of our posterior belief. In contrast to sample-based posterior approximations, our representation can capture a much richer space of DAGs, while also being able to tractably reason about the uncertainty through a range of useful inference queries. We empirically show how probabilistic circuits can be used as an augmented representation for structure learning methods, leading to improvement in both the quality of inferred structures and posterior uncertainty. Experimental results on conditional query answering further demonstrate the practical utility of the representational capacity of TRUST.