Theory and Practice of Logic Programming

Regular Papers

CHR(PRISM)-based probabilistic logic learning

JON SNEYERSa1, WANNES MEERTa1, JOOST VENNEKENSa1, YOSHITAKA KAMEYAa2 and TAISUKE SATOa2

a1 Department of Computer Science, K.U. Leuven, Belgium (e-mail: jon.sneyers@cs.kuleuven.be, wannes.meert@cs.kuleuven.be, joost.vennekens@cs.kuleuven.be)

a2 Tokyo Institute of Technology, Japan (e-mail: kameya@mi.cs.titech.ac.jp, sato@mi.cs.titech.ac.jp)

Abstract

PRISM is an extension of Prolog with probabilistic predicates and built-in support for expectation-maximization learning. Constraint Handling Rules (CHR) is a high-level programming language based on multi-headed multiset rewrite rules.

In this paper, we introduce a new probabilistic logic formalism, called CHRiSM, based on a combination of CHR and PRISM. It can be used for high-level rapid prototyping of complex statistical models by means of “chance rules”. The underlying PRISM system can then be used for several probabilistic inference tasks, including probability computation and parameter learning. We define the CHRiSM language in terms of syntax and operational semantics, and illustrate it with examples. We define the notion of ambiguous programs and define a distribution semantics for unambiguous programs. Next, we describe an implementation of CHRiSM, based on CHR(PRISM). We discuss the relation between CHRiSM and other probabilistic logic programming languages, in particular PCHR. Finally, we identify potential application domains.

(Received February 05 2010)

(Revised April 21 2010)

(Accepted May 12 2010)

KEYWORDS:

  • probabilistic logic learning;
  • constraint handling rules