Abstract
This paper presents an algorithm for inducing recursive first order Horn clause programs from examples without background knowledge. This algorithm invents new predicates and their definitions exhaustively until the instances of a new predicate become the same as examples except for the name of the predicate. Our system CIRP switches into constructive induction mode using a new heuristic taking advantage of the goal directed usefulness of incomplete clauses and of the fact that it is supplied with no background knowledge. It enables CIRP to avoid exhaustive search and to overcome some difficulties associated with employing encoding length principle as a switching element for constructive induction. This paper also describes a method for deciding the argument set for a new predicate by employing the structure of the arguments of the original predicate and reports the scope, limitation and remedy of limitation of this method.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
F. Bergadano and D. Gunetti. An interactive system to learn functional logic programs. In Procedings of the 13th IJCAI, 1993.
B. Kijsirikul, M. Numao, and M. Shimura. Efficient learning of logic programs with non-determinate, non-discriminating literals. In 8th Intl. Workshop on Machine Learning, 1991.
B. Kijsirikul, M. Numao, and M. Shimura. Discrimination-based constructive induction. In AAAI, 1992.
S. Lapointe, C. Ling, and S. Matwin. Constructive inductive logic programming. In International Joint Conference on Artificial Intelligence, 1993.
S. Lapointe and S. Matwin. Sub-unification: A tool for efficient induction of recursive programs. In 9th Intl. Workshop on Machine Learning, 1992.
S. Muggleton and W. Buntine. Machine invention of first-order predicates by inverting resolution. In 5th Intl. Workshop on Machine Learning, 1988.
S. Muggleton and C. Feng. Efficient induction of logic programs. In Procedings of the. 1st International workshop on Algorithmic Learning Theory, 1990.
J. R. Quinlan. Induction of decision trees. Machine Learning, 1 (1), 1986.
J. R. Quinlan. Learning logical definitions from relations. Machine Learning, 5:239–266, 1990.
L. De Raedt and M. Bruynooghe. Interactive concept-learning and constructive induction by analogy. Machine Learning, 8(2):107–150, 1992.
C. Rouveirol. Extensions of Inversion of Resolution Applied to Theory Completiion. Inductive Logic Programming, Academic Press, 1992.
Ehud Shapiro. Algorithmic Program Debugging. MIT Press, 1982.
G. Silverstein and M. J. Pazzani. Relational cliches: Constraining constructive induction during relational learning. In Procedings of Machine Learning Workshop, 1991.
R. Wirth. Learning by failure to prove. In Proceedings of EWSL, 1988.
R. Wirth and P. O'Rorke. Constraints for Predicate Invention. Inductive Logic Programming, Academic Press, 1992.
J. Wogulis and P. Langley. Improving efficiency by learning intermediate concepts. In Proceedings of the 11th IJCAI, 1989.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1994 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Mofizur, C.R., Numao, M. (1994). Constructive induction for recursive programs. In: Arikawa, S., Jantke, K.P. (eds) Algorithmic Learning Theory. AII ALT 1994 1994. Lecture Notes in Computer Science, vol 872. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-58520-6_62
Download citation
DOI: https://doi.org/10.1007/3-540-58520-6_62
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-58520-6
Online ISBN: 978-3-540-49030-2
eBook Packages: Springer Book Archive