Inference and learning in probabilistic logic programs using weighted Boolean formulas

DOI10.1017/S1471068414000076zbMATH Open1379.68062arXiv1304.6810MaRDI QIDQ4592983FDOQ4592983

Authors: Daan Fierens, Guy Van den Broeck, Joris Renkens, Dimitar Shterionov, Bernd Gutmann, Ingo Thon, Gerda Janssens, Luc De Raedt 

Publication date: 9 November 2017

Published in: Theory and Practice of Logic Programming (Search for Journal in Brave)

Abstract: Probabilistic logic programs are logic programs in which some of the facts are annotated with probabilities. This paper investigates how classical inference and learning tasks known from the graphical model community can be tackled for probabilistic logic programs. Several such tasks such as computing the marginals given evidence and learning from (partial) interpretations have not really been addressed for probabilistic logic programs before. The first contribution of this paper is a suite of efficient algorithms for various inference tasks. It is based on a conversion of the program and the queries and evidence to a weighted Boolean formula. This allows us to reduce the inference tasks to well-studied tasks such as weighted model counting, which can be solved using state-of-the-art methods known from the graphical model and knowledge compilation literature. The second contribution is an algorithm for parameter estimation in the learning from interpretations setting. The algorithm employs Expectation Maximization, and is built on top of the developed inference algorithms. The proposed approach is experimentally evaluated. The results show that the inference algorithms improve upon the state-of-the-art in probabilistic logic programming and that it is indeed possible to learn the parameters of a probabilistic logic program from interpretations.

Full work available at URL: https://arxiv.org/abs/1304.6810

Recommendations

• MAP inference for probabilistic logic programming
• Approximate Inference for Logic Programs with Annotated Disjunctions
• Compacting Boolean formulae for inference in probabilistic logic programming
• Constraint-based inference in probabilistic logic programs
• Lifted discriminative learning of probabilistic logic programs

zbMATH Keywords

parameter learningprobabilistic inferenceprobabilistic logic programming

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Reasoning under uncertainty in the context of artificial intelligence (68T37) Logic programming (68N17)

Cites Work

• CP-logic: A language of causal probabilistic events and its relation to logic programming
• Well–definedness and efficient inference for probabilistic logic programming under the distribution semantics
• Graph-Based Algorithms for Boolean Function Manipulation
• Modeling and Reasoning with Bayesian Networks
• The well-founded semantics for general logic programs
• Title not available (Why is that?)
• Graph minors. II. Algorithmic aspects of tree-width
• Probabilistic inductive logic programming. Theory and applications
• Title not available (Why is that?)
• Title not available (Why is that?)
• On the Tractable Counting of Theory Models and its Application to Truth Maintenance and Belief Revision
• Title not available (Why is that?)
• On the implementation of the probabilistic logic programming language ProbLog
• Compiling relational Bayesian networks for exact inference
• Logical and Relational Learning
• CP-logic theory inference with contextual variable elimination and comparison to BDD based inference methods
• Logic programming revisited: logic programs as inductive definitions
• Dedicated tabling for a probabilistic setting

Cited In (68)

• IASCAR: incremental answer set counting by anytime refinement
• Aspmc: new frontiers of algebraic answer set counting
• Using SWISH to realize interactive web-based tutorials for logic-based languages
• Semirings for probabilistic and neuro-symbolic logic programming
• Towards an effective practice of learning from data and knowledge
• Unified decomposition-aggregation (UDA) rules: dynamic, schematic, novel axioms
• Rule Induction and Reasoning over Knowledge Graphs
• Languages for probabilistic modeling over structured and relational domains
• Generating random instances of weighted model counting. An empirical analysis with varying primal treewidth
• Lifted Bayesian filtering in multiset rewriting systems
• Disjunctive delimited control
• Declarative probabilistic logic programming in discrete-continuous domains
• Generative Datalog and answer set programming (extended abstract)
• Generative Datalog with continuous distributions
• Statistical relational extension of answer set programming
• P-log: refinement and a new coherency condition
• Exploiting local and repeated structure in dynamic Bayesian networks
• Learning to rank in PRISM
• A taxonomy of weight learning methods for statistical relational learning
• Model checking finite-horizon Markov chains with probabilistic inference
• Closed-form solutions in learning probabilistic logic programs by exact score maximization
• Speeding up parameter and rule learning for acyclic probabilistic logic programs
• Exact stochastic constraint optimisation with applications in network analysis
• A functional account of probabilistic programming with possible worlds. Declarative pearl
• Weighted model counting without parameter variables
• Lifted inference with tree axioms
• Title not available (Why is that?)
• Probabilistic abstract argumentation frameworks, a possible world view
• Symbolic DNN-tuner
• Interpretable encoding of densities using possibilistic logic
• Knowledge compilation of logic programs using approximation fixpoint theory
• Compacting Boolean formulae for inference in probabilistic logic programming
• Utilizing treewidth for quantitative reasoning on epistemic logic programs
• \(T_{\mathcal{P}}\)-compilation for inference in probabilistic logic programs
• Dynamic Programming Algorithms as Products of Weighted Logic Programs
• Optimizing probabilities in probabilistic logic programs
• Diffusion centrality: a paradigm to maximize spread in social networks
• Probabilistic abductive logic programming using Dirichlet priors
• Explanations as programs in probabilistic logic programming
• Handling epistemic and aleatory uncertainties in probabilistic circuits
• Bandit-based Monte-Carlo structure learning of probabilistic logic programs
• Efficient inference and learning in a large knowledge base. Reasoning with extracted information using a locally groundable first-order probabilistic logic
• Probabilistic (logic) programming concepts
• Confidences for commonsense reasoning
• Weighted positive binary decision diagrams for exact probabilistic inference
• A survey of lifted inference approaches for probabilistic logic programming under the distribution semantics
• Efficient Knowledge Compilation Beyond Weighted Model Counting
• Neural probabilistic logic programming in DeepProbLog
• Connecting knowledge compilation classes and width parameters
• Some thoughts on knowledge-enhanced machine learning
• Probabilistic sentence satisfiability: an approach to PSAT
• Advanced SMT techniques for weighted model integration
• Constraint-based inference in probabilistic logic programs
• Learning hierarchical probabilistic logic programs
• Products of weighted logic programs
• Computing \(\mathrm{LP}^{\mathrm{MLN}}\) using ASP and MLN solvers
• MAP inference for probabilistic logic programming
• Online event recognition over noisy data streams
• IASCAR: incremental answer set counting by anytime refinement
• Predictive spreadsheet autocompletion with constraints
• Approximate weighted model integration on DNF structures
• The joy of probabilistic answer set programming: semantics, complexity, expressivity, inference
• Thirty years of credal networks: specification, algorithms and complexity
• The complexity of Bayesian networks specified by propositional and relational languages
• Connecting width and structure in knowledge compilation
• On the complexity of propositional and relational credal networks
• The finite model theory of Bayesian network specifications: descriptive complexity and zero/one laws
• Complexity results for probabilistic answer set programming

Uses Software

• ASSAT
• ProbLog
• CP-logic
• Dsharp

This page was built for publication: Inference and learning in probabilistic logic programs using weighted Boolean formulas

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4592983)