Bayesian Inference by Symbolic Model Checking
This work addresses the challenge of scalable inference in Bayesian networks for AI and probabilistic modeling, but it is incremental as it builds on existing model checking and symbolic methods.
The paper tackles the problem of performing inference in Bayesian networks by translating them into tree-like Markov chains and using probabilistic model checking to compute reachability probabilities, resulting in an implementation that shows symbolic data structures like MTBDDs are effective on large benchmarks, with comparisons to existing symbolic techniques.
This paper applies probabilistic model checking techniques for discrete Markov chains to inference in Bayesian networks. We present a simple translation from Bayesian networks into tree-like Markov chains such that inference can be reduced to computing reachability probabilities. Using a prototypical implementation on top of the Storm model checker, we show that symbolic data structures such as multi-terminal BDDs (MTBDDs) are very effective to perform inference on large Bayesian network benchmarks. We compare our result to inference using probabilistic sentential decision diagrams and vtrees, a scalable symbolic technique in AI inference tools.