Abstract
In this paper, we close the gap between the simple and straight-forwardimplemen tations of top-down hill-climbing that can be foundin the literature, andthe rather complex strategies for greedy bottom-up generalization. Our main result is that the simple bottom-up counterpart to the top-down hill-climbing algorithm is unable to learn in domains with dispersed examples. In particular, we show that guided greedy generalization is impossible if the seed example differs in more than one attribute value from its nearest neighbor. We also perform an empirical study of the commonness of this problem is in popular benchmark datasets, andpresen t average-case andw orst-case results for the probability of drawing a pathological seed example in binary domains.
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
Zhang] F. Bergadano, S. Matwin, R. S. Michalski, and J. Zhang. Learning twotieredd escriptions of flexible concepts: The POSEIDON system. Machine Learning, 8:5–43, 1992.
C. L. Blake and C. J. Merz. UCI repository of machine learning databases. http://www.ics.uci.edu/~mlearn/MLRepository.html, 1998. Department of Information andComputer Science, University of California at Irvine, Irvine CA.
Y. Chevaleyre and J.-D. Zucker. A framework for learning rules from multiple instance data. In L. D. Raedt and P. Flach, editors, Proceedings ofthe 12th European Conference on Machine Learning (ECML-01), pages 49–60, Freiburg, Germany, 2001. Springer-Verlag.
P. Clark and R. Boswell. Rule induction with CN2: Some recent improvements. In Proceedings ofthe 5th European Working Session on Learning (EWSL-91), pages 151–163, Porto, Portugal, 1991. Springer-Verlag.
P. Clark and T. Niblett. The CN2 induction algorithm. Machine Learning, 3(4): 261–283, 1989.
W. W. Cohen. Fast effective rule induction. In A. Prieditis and S. Russell, editors, Proceedings ofthe 12th International Conference on Machine Learning (ML-95), pages 115–123, Lake Tahoe, CA, 1995. Morgan Kaufmann.
L. De Raedt. Attribute value learning versus inductive logic programming: The missing links (extended abstract). In D. Page, editor, Proceedings ofthe 8th International Conference on Inductive Logic Programming (ILP-98), pages 1–8. Springer-Verlag, 1998.
L. De Raedt and W. Van Laer. Inductive constraint logic. In Proceedings ofthe 5th Workshop on Algorithmic Learning Theory (ALT-95). Springer-Verlag, 1995.
P. Domingos. Unifying instance-basedandrule-based induction. Machine Learning, 24:141–168, 1996.
S. Džeroski and I. Bratko. Handling noise in Inductive Logic Programming. In S. H. Muggleton and K. Furukawa, editors, Proceedings ofthe 2nd International Workshop on Inductive Logic Programming (ILP-92), number TM-1182 in ICOT Technical Memorandum, pages 109–125, Tokyo, Japan, 1992. Institue for New Generation Computer Technology.
J. Fürnkranz. Fossil: A robust relational learner. In F. Bergadano and L. De Raedt, editors, Proceedings ofthe 7th European Conference on Machine Learning (ECML-94), volume 784 of Lecture Notes in Artificial Intelligence, pages 122–137, Catania, Italy, 1994. Springer-Verlag.
J. Fürnkranz. Pruning algorithms for rule learning. Machine Learning, 27(2): 139–171, 1997.
J. Fürnkranz. Separate-and-conquer rule learning. Artificial Intelligence Review, 13(1):3–54, February 1999.
J. Fürnkranz and G. Widmer. Incremental Reduced Error Pruning. In W. Cohen and H. Hirsh, editors, Proceedings ofthe 11th International Conference on Machine Learning (ML-94), pages 70–77, New Brunswick, NJ, 1994. Morgan Kaufmann.
K. A. Kaufman and R. S. Michalski. An adjustable rule learner for pattern discovery using the AQ methodology. Journal ofIntelligent Information Systems, 14:199–216, 2000.
R. Kohavi and G.H. John. Wrappers for feature subset selection. Artificial Intelligence, 97(1-2):273–324, 1997. Special Issue on Relevance.
N. Lavrač, P. Flach, and B. Zupan. Rule evaluation measures: A unifying view. In S. Džeroski and P. Flach, editors, Ninth International Workshop on Inductive Logic Programming (ILP’99), volume 1634 of Lecture Notes in Artificial Intelligence, pages 174–185. Springer-Verlag, June 1999.
R. S. Michalski. On the quasi-minimal solution of the covering problem. In Proceedings of the 5th International Symposium on Information Processing (FCIP-69), volume A3 (Switching Circuits), pages 125–128, Bled, Yugoslavia, 1969.
R. S. Michalski. Pattern recognition andrule-guided inference. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2:349–361, 1980.
R. S. Michalski, I. Mozetič, J. Hong, and N. Lavrač. The multi-purpose incremental learning system AQ15 and its testing application to three medical domains. In Proceedings ofthe 5th National Conference on Artificial Intelligence (AAAI-86), pages 1041–1045, Philadelphia, PA, 1986.
S. H. Muggleton. Inverse entailment and Progol. New Generation Computing, 13 (3,4):245–286, 1995. Special Issue on Inductive Logic Programming.
S. H. Muggleton and C. Feng. Efficient induction of logic programs. In Proceedings ofthe 1st Conference on Algorithmic Learning Theory, pages 1–14, Tokyo, Japan, 1990.
J. R. Quinlan. Learning logical definitions from relations. Machine Learning, 5: 239–266, 1990.
J. R. Quinlan. Determinate literals in inductive logic programming. In Proceedings of the 8th International Workshop on Machine Learning (ML-91), pages 442–446, 1991.
J. R. Quinlan and R. M. Cameron-Jones. Induction of logic programs: FOIL and relatedsystems. New Generation Computing, 13(3,4):287–312, 1995. Special Issue on Inductive Logic Programming.
G. I. Webb. Learning disjunctive class descriptions by least generalisation. Technical Report TR C92/9, Deakin University, School of Computing & Mathematics, Geelong, Australia, September 1992.
G. I. Webb and J. W. M. Agar. Inducing diagnostic rules for glomerular disease with the DLG machine learning algorithm. Artificial Intelligence in Medicine, 4: 419–430, 1992.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Fürnkranz, J. (2002). A Pathology of Bottom-Up Hill-Climbing in Inductive Rule Learning. In: Cesa-Bianchi, N., Numao, M., Reischuk, R. (eds) Algorithmic Learning Theory. ALT 2002. Lecture Notes in Computer Science(), vol 2533. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-36169-3_22
Download citation
DOI: https://doi.org/10.1007/3-540-36169-3_22
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-00170-6
Online ISBN: 978-3-540-36169-5
eBook Packages: Springer Book Archive