Efficient Search-Based Weighted Model Integration

Abstract

Weighted model integration (WMI) extends weighted model counting to integration in mixed discrete-continuous domains. It has shown tremendous promise for solving probabilistic inference problems in graphical models and probabilistic programs. Yet, state-of-the-art tools for WMI have limited performance and ignore the independence structure that is crucial to improving efficiency. To address this limitation, we propose an efficient model integration algorithm for theories with tree pri-mal graphs. We exploit the sparse graph structure by using search to performing integration. Our algorithm greatly improves the computational efficiency on such problems and exploits context-specific independence between variables. Experimental results show dramatic speedups compared to existing WMI solvers on problems with tree-shaped dependencies.

Zhe Zeng

Assistant Professor

I do research in probabilistic ML and neurosymbolic AI to enable and support decision-making in the real world in the presence of probabilistic uncertainty and symbolic knowledge, where the symbolic knowledge can be graph structures and logical, arithmetic, and physical constraints.