Computing LPMLN using ASP and MLN solvers*. Issue 5 (30th August 2017)

• Record Type: Journal Article
• Authors: LEE, JOOHYUNG; TALSANIA, SAMIDH; WANG, YI
• Editors: Rocha, Ricardo; Cao Son, Tran
• Is Part Of: Theory and practice of logic programming. Volume 17:Issue 5/6(2017)
• Page Start: 942
• Page End: 960
• Publication Date: 2017-08-30
• Subjects: Answer Set Programming, -- Markov Logic, -- LPMLN; Logic programming -- Periodicals; Artificial intelligence -- Computer programs -- Periodicals; Constraint programming (Computer science) -- Periodicals; 005.115
• DOI: 10.1017/S1471068417000400
• Languages: English
• ISSNs: 1471-0684

Abstract

Abstract: LP MLN is a recent addition to probabilistic logic programming languages. Its main idea is to overcome the rigid nature of the stable model semantics by assigning a weight to each rule in a way similar to Markov Logic is defined. We present two implementations of LP MLN, lpmln 2asp andlpmln 2mln . Systemlpmln 2asp translates LP MLN programs into the input language of answer set solverclingo, and using weak constraints and stable model enumeration, it can compute most probable stable models as well as exact conditional and marginal probabilities. Systemlpmln 2mln translates LP MLN programs into the input language of Markov Logic solvers, such asalchemy, tuffy, androckit, and allows for performing approximate probabilistic inference on LP MLN programs. We also demonstrate the usefulness of the LP MLN systems for computing other languages, such as ProbLog and Pearl's Causal Models, that are shown to be translatable into LP MLN .