MAP inference for probabilistic logic programming
From MaRDI portal
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.
Recommendations
- Constraint-based inference in probabilistic logic programs
- The complexity of inferences and explanations in probabilistic logic programming
- Inference and learning in probabilistic logic programs using weighted Boolean formulas
- Lifted variable elimination for probabilistic logic programming
- Focused most probable world computations in probabilistic logic programs
Cites work
- scientific article; zbMATH DE number 1946853 (Why is no real title available?)
- scientific article; zbMATH DE number 7102019 (Why is no real title available?)
- Abducing through negation as failure: stable models within the independent choice logic
- Efficient manipulation of decision diagrams
- Emergence of Scaling in Random Networks
- Inference and learning in probabilistic logic programs using weighted Boolean formulas
- Logic Programming
- Probabilistic inductive logic programming. Theory and applications
- Tabling and answer subsumption for reasoning on logic programs with annotated disjunctions
- The Complexity of Enumeration and Reliability Problems
- The PITA system: tabling and answer subsumption for reasoning under uncertainty
- The distribution semantics for normal programs with function symbols
- The independent choice logic for modelling multiple agents under uncertainty
- Well–definedness and efficient inference for probabilistic logic programming under the distribution semantics
Cited in
(16)- Evaluating inference algorithms for the Prolog factor language
- Aspmc: new frontiers of algebraic answer set counting
- Inference and learning in probabilistic logic programs using weighted Boolean formulas
- Probabilistic abductive logic programming using Dirichlet priors
- The complexity of inferences and explanations in probabilistic logic programming
- Efficient Knowledge Compilation Beyond Weighted Model Counting
- Using histograms to better answer queries to probabilistic logic programs
- Computing most probable worlds of action probabilistic logic programs: scalable estimation for \(10^{30,000}\) worlds
- Lifted variable elimination for probabilistic logic programming
- Constraint-based inference in probabilistic logic programs
- Nonground abductive logic programming with probabilistic integrity constraints
- Abduction with probabilistic logic programming under the distribution semantics
- Online event recognition over noisy data streams
- The PITA system: tabling and answer subsumption for reasoning under uncertainty
- Causal inference in cplint
- k-optimal: a novel approximate inference algorithm for ProbLog
This page was built for publication: MAP inference for probabilistic logic programming
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5140005)
