Inductive Constraint Logic Programming: An Overview

. This paper provides a brief introduction and overview of the emerging area of Inductive Constraint Logic Programming (ICLP). It discusses some of the existing work in the area and presents some of the research issues and open questions that need to be addressed. 1 Introduction Inductive Logic Programming (ILP) refers to a class of machine learning algorithms where the agent learns a first-order theory from examples and background knowledge. The ILP framework in machine learning is perhaps the most general of all because of the complexity of the concepts learned. The use of first-order logic programs as the underlying representation makes ILP systems more powerful and useful than the conventional empirical machine learning systems. ILP systems have been successfully used in a variety of real life domains including mesh design, protein synthesis, games and fault diagnosis. Most of the first-order representations used in ILP systems are variants of horn-clause based clausal logic o..