Abstract
In this paper we present and prove both negative and positive theoretical results concerning the representation and evaluation of first-order logic clauses using genetic algorithms. Over the last few years, a few approaches have been proposed aiming to combine genetic and evolutionary computation (EC) with inductive logic programming (ILP). The underlying rationale is that evolutionary algorithms, such as genetic algorithms, might mitigate the combinatorial explosions generated by the inductive learning of rich representations, such as those used in first-order logic. Particularly, the genetic algorithms approach to ILP presented by Tamaddoni-Nezhad and Muggleton has attracted the attention of both the EC and ILP communities in recent years. Unfortunately, a series of systematic and fundamental theoretical errors renders their framework moot. This paper critically examines the fallacious claims in the mentioned approach. It is shown that, far from restoring completeness to the learner progol‘s search of the subsumption lattice, their proposed binary representation is both incomplete and noncompact. It is also shown that their fast evaluation of clauses based on bitwise operations is based on false theorems and therefore invalid. As an alternative to Tamaddoni-Nezhad and Muggleton’s flawed genetic algorithms attempt, we propose a framework based on provably sound and complete binary refinement.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Alphonse, É., Rouveirol, C.: Object Identity for Relational Learning. Technical report, LRI, Université Paris-Sud, Supported by ESPRIT Framework IV through LTR ILP2 (1999)
Alphonse, É., Rouveirol, C.: Test Incorporation for Propositionalization Methods in ILP. Technical report, LRI, Université Paris-Sud, Supported by ESPRIT Framework IV through LTR ILP2 (1999)
Anglano, C., Giordana, A., Bello, G.L., Saitta, L.: An Experimental Evaluation of Coevolutive Concept Learning. In: Shavlik, J. (ed.) Proceedings of the Fifteenth International Conference on Machine Learning, pp. 19–27. Morgan Kaufmann, San Francisco (1998)
Badea, L., Stanciu, M.: Refinement Operators Can Be (Weakly) Perfect. In: Džeroski, S., Flach, P.A. (eds.) ILP 1999. LNCS (LNAI), vol. 1634, pp. 21–32. Springer, Heidelberg (1999)
Cussens, J., Frisch, A. (eds.) Inductive Logic Programming—ILP 2000, Proceedings of the 10th International Conference on Inductive Logic Programming, Imperial College, UK (July 2000) (work-in-progress reports)
Cussens, J., Frisch, A. (eds.) Inductive Logic Programming—ILP 2000, Proceedings of the 10th International Conference on Inductive Logic Programming, Imperial College, UK. LNCS (LNAI), vol. 1866, Springer, Heidelberg (2000)
De Raedt, L.: Attribute-Value Learning versus Inductive Logic Programming: The Missing Links. In: Page [23], pp. 1–8
Divina, F., Marchiori, E.: Knowledge Based Evolutionary Programming for Inductive Learning in First-Order Logic. In: Spector, L., et al. (eds.) [25], pp. 173–181
Furukawa, K., Ozaki, T.: On the Completion of Inverse Entailment for Mutual Recursion and its Application to Self Recursion. In: Cussens, Frisch [5], pp. 107–119
Giordana, A., Neri, F.: Search-Intensive Concept Induction. Evolutionary Computation Journal 3(4), 375–416 (1996)
Giordana, A., Sale, C.: Learning Structured Concepts Using Genetic Algorithms. In: Sleeman, D., Edwards, P. (eds.) Proceedings of the Ninth International Workshop on Machine Learning, pp. 169–178. Morgan Kaufmann, San Francisco (1992)
Hekanaho, J.: DOGMA: A GA-based Relational Learner. In: Page [23], pp. 205–214
Inoue, K.: Induction, Abduction and Consequence-Finding. In: Rouveirol, C., Sebag, M. (eds.) ILP 2001. LNCS (LNAI), vol. 2157, pp. 65–79. Springer, Heidelberg (2001)
Ito, K., Yamamoto, A.: Finding Hypotheses from Examples by Computing the Least Generalization of Bottom Clauses. In: Arikawa, S., Motoda, H. (eds.) DS 1998. LNCS (LNAI), vol. 1532, pp. 303–314. Springer, Heidelberg (1998)
Laird, P.D.: Learning from Good Data and Bad. PhD thesis, Yale University (1987)
Mărginean, F.A.: Combinatorics of Refinement. PhD thesis, Department of Computer Science, The University of York (September 2001)
Muggleton, S.: Inverse entailment and Progol. New Generation Computing 13, 245–286 (1995)
Muggleton, S.: Completing Inverse Entailment. In: Page [23], pp. 245–249
Muggleton, S.H., Mărginean, F.A.: Binary Refinement. In: McCarthy, J., Minker, J. (eds.) Workshop on Logic-Based Artificial Intelligence, College Park, Washington DC, Maryland, June 14–16. Computer Science Department, University of Maryland (1999)
Nienhuys-Cheng, S.-H., de Wolf, R.: Foundations of Inductive Logic Programming. LNCS (LNAI), vol. 1228. Springer, Heidelberg (1997)
Ohara, K., Babaguchi, N., Kitahashi, T.: An Efficient Hypothesis Search Algorithm Based on Best-Bound Strategy. In: Cussens, Frisch [5], pp. 212–225
Ohwada, H., Nishiyama, H., Mizoguchi, F.: Concurrent Execution of Optimal Hypothesis Search for Inverse Entailment. In: Cussens, Frisch [6], pp. 165–173
Page, D. (ed.) ILP 1998. LNCS (LNAI), vol. 1446. Springer, Heidelberg (1998)
Page, D.: ILP: Just Do It. In: Cussens, Frisch [6], pp. 25–40
Spector, L., Goodman, E.D., Wu, A., Langdon, W.B., Voigt, H.-M., Gen, M., Sen, S., Dorigo, M., Pezeshk, S., Garzon, M.H., Burke, E. (eds.) Proceedings of the Genetic and Evolutionary Computation Conference, GECCO-2001, San Francisco, CA, July 7–11, AAAI, Morgan Kaufmann (2001)
Tamaddoni-Nezhad, A., Muggleton, S.: A Genetic Algorithms Approach to ILP. In: Matwin, S., Sammut, C. (eds.) ILP 2002. LNCS (LNAI), vol. 2583, pp. 285–300. Springer, Heidelberg (2003)
Tamaddoni-Nezhad, A., Muggleton, S.H.: Searching the Subsumption Lattice by a Genetic Algorithm. In: Cussens, Frisch [6], pp. 243–252
Tamaddoni-Nezhad, A., Muggleton, S.H.: Using Genetic Algorithms for Learning Clauses in First-Order Logic. In: Spector et al [25], pp. 639–646
Torre, F., Rouveirol, C.: Private Properties and Natural Relations in Inductive Logic Programming. Technical report, LRI, Université Paris-Sud (July 1997)
van der Laag, P.R.: An Analysis of Refinement Operators in Inductive Logic Programming. Technical Report 102, Tinbergen Institute Research Series (1995)
Yamamoto, A.: Which Hypotheses Can Be Found with Inverse Entailment? In: Džeroski, S., Lavrač, N. (eds.) ILP 1997. LNCS, vol. 1297, pp. 296–308. Springer, Heidelberg (1997)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Mărginean, F.A. (2003). Which First-Order Logic Clauses Can Be Learned Using Genetic Algorithms?. In: Horváth, T., Yamamoto, A. (eds) Inductive Logic Programming. ILP 2003. Lecture Notes in Computer Science(), vol 2835. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-39917-9_16
Download citation
DOI: https://doi.org/10.1007/978-3-540-39917-9_16
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-20144-1
Online ISBN: 978-3-540-39917-9
eBook Packages: Springer Book Archive