TP-Compilation for inference in probabilistic logic programs

Abstract

We propose T P -compilation, a new inference technique for probabilistic logic programs that is based on forward reasoning. T P -compilation proceeds incrementally in that it interleaves the knowledge compilation step for weighted model counting with forward reasoning on the logic program. This leads to a novel anytime algorithm that provides hard bounds on the inferred probabilities. The main difference with existing inference techniques for probabilistic logic programs is that these are a sequence of isolated transformations. Typically, these transformations include conversion of the ground program into an equivalent propositional formula and compilation of this formula into a more tractable target representation for weighted model counting. An empirical evaluation shows that T P -compilation effectively handles larger instances of complex or cyclic real-world problems than current sequential approaches, both for exact and anytime approximate inference. Furthermore, we show that T P -compilation is conducive to inference in dynamic domains as it supports efficient updates to the compiled model.

onto2problog: A Probabilistic Ontology-Mediated Querying System using Probabilistic Logic Programming

“…We used two different inference methods supported by ProbLog: (1) the “classic” ProbLog inference approach of cycle-breaking and compilation to sentential decision diagrams (SDDs) [ 21 ], and (2) -compilation to SDDs, which avoids the cycle-breaking step altogether through forward inference [ 22 ]. Regardless of the method used, ProbLog first computes the ground program relevant to the query, that is, it transforms the probabilistic logic program into one using only ground atoms (while returning the same probabilities).…”

“…ProbLog supports marginal inference via a variety of different algorithms based on knowledge compilation [ 6 ], for example, to d-DNNF and SDD. It also supports forward inference in a process known as -compilation [ 22 ]. Using ProbLog’s Python interface, the user may select which inference method they wish to use in order to evaluate their query.…”

Beyond the Grounding Bottleneck: Datalog Techniques for Inference in Probabilistic Logic Programs

“…Furthermore, for ten of the queries, vProbLog computes exact answers over 1M facts in seconds. At the same time, on three standard PLP benchmarks (Fierens et al 2015;Renkens et al 2014;Vlasselaer et al 2016) where the bottleneck is formula construction, vProbLog achieves comparable approximations to the existing implementation in less time.…”

“…Our key contribution is a program transformation approach that allows us to implement forward inference using an efficient Datalog engine that directly operates on non-ground functorfree programs. We focus on programs without negation for simplicity, though the Tc P operator has been studied for general probabilistic logic programs (Bogaerts and Van den Broeck 2015;Riguzzi 2016) as well; the extension to stratified negation following (Vlasselaer et al 2016) is straightforward. We further build upon two well-known techniques from the Datalog community, namely semi-naive evaluation (Abiteboul, Hull, and Vianu 1995), which avoids recomputing the same consequences repeatedly during forward reasoning, and the magic sets transformation (Bancilhon et al 1986;Beeri and Ramakrishnan 1991), which makes forward reasoning query driven.…”

“…While λ may use variables besides the facts, e.g., variables corresponding to the atoms in the program, their values have to be uniquely defined for each total choice. In what follows we present (and adapt) notions and results by Vlasselaer et al (2016).…”