Rao-Blackwellising Bayesian Causal Inference

arXiv:2402.14781v1 Announce Type: new
Abstract: Bayesian causal inference, i.e., inferring a posterior over causal models for the use in downstream causal reasoning tasks, poses a hard computational inference problem that is little explored in literature. In this work, we combine techniques from order-based MCMC structure learning with recent advances in gradient-based graph learning into an effective Bayesian causal inference framework. Specifically, we decompose the problem of inferring the causal structure into (i) inferring a topological order over variables and (ii) …

abstract advances arxiv bayesian causal inference computational cs.ai cs.lg gradient graph graph learning inference literature mcmc posterior reasoning stat.me stat.ml tasks type work