Efficient Search-Based Weighted Model Integration
This work addresses efficiency issues in inference for graphical models and probabilistic programming, though it is incremental as it focuses on a specific graph structure.
The paper tackled the performance limitations of weighted model integration (WMI) tools by proposing an efficient algorithm for theories with tree primal graphs, resulting in dramatic speedups compared to existing solvers on such problems.
Weighted model integration (WMI) extends Weighted model counting (WMC) to the integration of functions over mixed discrete-continuous domains. It has shown tremendous promise for solving inference problems in graphical models and probabilistic programming. Yet, state-of-the-art tools for WMI are limited in terms of 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 primal 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.