Paper Info
Title
Probabilistic termination of CHRiSM programs
Abstract
Termination analysis has received considerable attention in Logic Programming for several decades. In recent years, probabilistic extensions of Logic Programming languages have become increasingly important. Languages like PRISM, CP-Logic, ProbLog, and CHRiSM have been introduced and proved very useful for addressing problems in which a combination of logical and probabilistic reasoning is required. As far as we know, the termination of probabilistic logical programs has not received any attention in the community so far. Termination of a probabilistic program is not a crisp notion. Given a query, such a program does not simply either terminate or not terminate, but it terminates with a certain probability. In this paper, we explore this problem in the context of CHRiSM, a probabilistic extension of CHR. We formally introduce the notion of probabilistic termination. We study this concept on the basis of a number of case studies. We provide some initial sufficient conditions to characterize probabilistically terminating programs and queries. We also discuss some challenging examples that reveal the complexity and interest of more general settings. The paper is intended as a first step in a challenging and important new area in the analysis of Logic Programs.
Year
DOI
Venue
2011
10.1007/978-3-642-32211-2_15
LOPSTR
Keywords
Field
DocType
challenging example,probabilistic program,logic programming language,chrism program,probabilistic extension,logic programs,logic programming,probabilistic logical program,termination analysis,probabilistic reasoning,probabilistic termination
Logical conjunction,Computer science,Algorithm,Theoretical computer science,Termination analysis,Logic programming,Probabilistic argumentation,Probabilistic logic,Constraint Handling Rules
Conference
Citations 
PageRank 
References 
4
0.51
26
Authors
2
Name
Order
Citations
PageRank
jon sneyers11179.47
Danny de Schreye2135275.10