Reformulating Inference Problems Through Selective Conditioning
This work addresses computational bottlenecks in probabilistic inference for AI and machine learning applications, but it appears incremental as it builds on existing BNRAS algorithms and selective conditioning methods.
The paper tackles the problem of intractable probabilistic inference in belief networks by selectively conditioning on specific nodes to decompose the problem into multiple tractable subproblems, using stochastic-simulation algorithms like BNRAS and logic sampling, and analyzes computational tradeoffs.
We describe how we selectively reformulate portions of a belief network that pose difficulties for solution with a stochastic-simulation algorithm. With employ the selective conditioning approach to target specific nodes in a belief network for decomposition, based on the contribution the nodes make to the tractability of stochastic simulation. We review previous work on BNRAS algorithms- randomized approximation algorithms for probabilistic inference. We show how selective conditioning can be employed to reformulate a single BNRAS problem into multiple tractable BNRAS simulation problems. We discuss how we can use another simulation algorithm-logic sampling-to solve a component of the inference problem that provides a means for knitting the solutions of individual subproblems into a final result. Finally, we analyze tradeoffs among the computational subtasks associated with the selective conditioning approach to reformulation.