MAP Inference for Probabilistic Logic Programming
Abstract
In Probabilistic Logic Programming (PLP) the most commonly studied inference task is to compute the marginal probability of a query given a program. In this paper, we consider two other important tasks in the PLP setting: the Maximum-A-Posteriori (MAP) inference task, which determines the most likely values for a subset of the random variables given evidence on other variables, and the Most Probable Explanation (MPE) task, the instance of MAP where the query variables are the complement of the evidence variables. We present a novel algorithm, included in the PITA reasoner, which tackles these tasks by representing each problem as a Binary Decision Diagram and applying a dynamic programming procedure on it. We compare our algorithm with the version of ProbLog that admits annotated disjunctions and can perform MAP and MPE inference. Experiments on several synthetic datasets show that PITA outperforms ProbLog in many cases.
Cite
@article{arxiv.2008.01394,
title = {MAP Inference for Probabilistic Logic Programming},
author = {Elena Bellodi and Marco Alberti and Fabrizio Riguzzi and Riccardo Zese},
journal= {arXiv preprint arXiv:2008.01394},
year = {2020}
}
Comments
Paper presented at the 36th International Conference on Logic Programming (ICLP 2020), University Of Calabria, Rende (CS), Italy, September 2020, 16 pages