Probabilistic Circuits with Constraints via Convex Optimization

arXiv:2403.13125v1 Announce Type: new
Abstract: This work addresses integrating probabilistic propositional logic constraints into the distribution encoded by a probabilistic circuit (PC). PCs are a class of tractable models that allow efficient computations (such as conditional and marginal probabilities) while achieving state-of-the-art performance in some domains. The proposed approach takes both a PC and constraints as inputs, and outputs a new PC that satisfies the constraints. This is done efficiently via convex optimization without the need to retrain the entire …

abstract art arxiv circuits class constraints cs.ai cs.lg distribution domains logic optimization pcs performance state tractable type via work