Knowledge Compilation with Continuous Random Variables and its Application in Hybrid Probabilistic Logic Programming
This work addresses a bottleneck in probabilistic reasoning for AI and logic programming communities, offering a novel solution for hybrid inference.
The paper tackles the challenge of exact probabilistic inference in hybrid domains with both discrete and continuous random variables, introducing a method based on weighted model integration and algebraic model counting that avoids trade-offs, and applies it to a new logic programming language called HAL-ProbLog.
In probabilistic reasoning, the traditionally discrete domain has been elevated to the hybrid domain encompassing additionally continuous random variables. Inference in the hybrid domain, however, usually necessitates to condone trade-offs on either the inference on discrete or continuous random variables. We introduce a novel approach based on weighted model integration and algebraic model counting that circumvents these trade-offs. We then show how it supports knowledge compilation and exact probabilistic inference. Moreover, we introduce the hybrid probabilistic logic programming language HAL-ProbLog, an extension of ProbLog, to which we apply our inference approach.