Elements of functional programming
Elements of functional programming
The role(s) of logic in artificial intelligence
Handbook of logic in artificial intelligence and logic programming (vol. 1)
Algebra of programming
Towards a logic programming methodology based on higher-order predicates
New Generation Computing
Logic Program Synthesis from Incomplete Information: By Pierre Flener
Logic Program Synthesis from Incomplete Information: By Pierre Flener
Foundations of Inductive Logic Programming
Foundations of Inductive Logic Programming
Selected Papers from the 6th International Workshop on Inductive Logic Programming
ILP '96 Selected Papers from the 6th International Workshop on Inductive Logic Programming
Inductive Logic Program Synthesis with DIALOGS
ILP '96 Selected Papers from the 6th International Workshop on Inductive Logic Programming
Ontological constitutions for classes and properties
ICCS'06 Proceedings of the 14th international conference on Conceptual Structures: inspiration and Application
| Hi-index | 0.00 |
This paper presents an approach to inductive synthesis of logicprograms from examples using problem decomposition and problemreduction principles. This is in contrast to the prevailing logicprogram induction paradigm, which relies on generalization ofprograms from examples. The problem reduction is accomplished as aconstrained top-down search process, which eventually is to reachtrivial problems.Our induction scheme applies a distinguished logic programminglanguage in which programs are combined from elementary predicatesby means of combinators conceived of as problem reduction operatorsincluding list recursion forms. The operator form admits inductivesynthesis as a top-down piecewise composition of semanticallymeaningful program elements according to the compositionalsemantics principle and with appeals neither to specialgeneralization mechanisms nor to alternative forms of resolutionand unification, or predicate invention.The search space is reduced by subjecting the induction processto various constraints concerning syntactical form, modes, datatypes, and computational resources. This is illustrated in thepaper with well-modedness constraints with the aim of synthesisingwell-moded, procedurally acceptable programs.Preliminary experiments with the proposed induction method leadus to tentatively conclude that the presented approach forms aviable alternative to the prevailing inductive logic programmingmethods applying generalization from examples.