Abstract
VS-CBR [14] is a simple instance-based learning algorithm that adjusts a weighted similarity measure as well as collecting cases. This paper presents a ‘PAC’ analysis of VS-CBR, motivated by the PAC learning framework, which demonstrates two main ideas relevant to the study of instance-based learners. Firstly, the hypothesis spaces of a learner on different target concepts can be compared to predict the difficulty of the target concepts for the learner. Secondly, it is helpful to consider the ‘constituent parts’ of an instance-based learner: to explore separately how many examples are needed to infer a good similarity measure and how many examples are needed for the case base. Applying these approaches, we show that VS-CBR learns quickly if most of the variables in the representation are irrelevant to the target concept and more slowly if there are more relevant variables. The paper relates this overall behaviour to the behaviour of the constituent parts of VS-CBR.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
D. W. Aha. Tolerating noisy, irrelevant and novel attributes in instance-based learning algorithms. International Journal of Man-Machine Studies, 36:267–287, 1992.
D. W. Aha, D. Kibler, and M. K. Albert. Instance-based learning algorithms. Machine Learning, 6:37–66, 1991.
M. Anthony and N. Biggs. Computational Learning Theory. Cambridge University Press, 1992.
A. Blumer, A. Ehrenfeucht, D. Haussler, and M. K. Warmuth. Learnability and the Vapnik-Chervonenkis dimension. Journal of the ACM, 36(4):929–965, Oct 1989.
J. P. Callan, T. E. Fawcett, and E. L. Rissland. CABOT: An adaptive approach to case-based search. In Proceedings of IJCAI-91, pp 803–808, Morgan Kaufmann, 1991.
C. Globig, K. P. Jantke, S. Lange, and Y. Sakakibara. On case-based learnability of languages. New Generation Computing, 15(1), 1997.
A. D. Griffiths. Inductive Generalisation in Case-Based Reasoning Systems. PhD thesis, Published as Technical Report YCST-97-02, Department of Computer Science, University of York, York YO1 5DD, UK, 1997.
K. P. Jantke. Case-based learning in inductive inference. In Proceedings of COLT92, pp 218–223. ACM Press, 1992.
P. Langley and W. Iba. Average-case analysis of a nearest neighbour algorithm. In R Bajcsy, editor, Proceedings of IJCAI-93, pp 889–894. Morgan Kaufmann, 1993.
T. M. Mitchell. Generalisation as search. Artificial Intelligence, 18(2):203–226, 1982.
K. Satoh and S. Okamoto. Towards PAC-learning of weights from qualitative distance information. In D W Aha, editor, Case-based reasoning: Papers from the 1994 AAAI Workshop, Technical Report WS-94-01. AAAI Press, 1994.
C. Stanfill and D. L. Waltz. Towards memory-based reasoning. Communications of the ACM, 29(12):1213–1228, 1986.
L. G. Valiant. Deductive learning. Philosophical Transactions of the Royal Philosophical Society of London A, 312:441–446, 1984.
S. Wess and C. Globig. Case-based and symbolic classification — A case study. In S Wess, K-D Althoff, and M M Richter, editors, Topics in GBR: Selected Papers from EWCBR93, LNCS 837, pp 77–91. Springer-Verlag, 1994.
D. Wettschereck, D. W. Aha, and T. Mohri. A review and comparative evaluation of feature weighting methods for lazy learning algorithms. Technical Report AIC-95-012, Navy Center for Applied Research in AI, Naval Research Laboratory, Washington, DC 20375-5337, USA, 1995.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Griffiths, A.D., Bridge, D.G. (1997). PAC analyses of a ‘similarity learning’ IBL algorithm. In: Leake, D.B., Plaza, E. (eds) Case-Based Reasoning Research and Development. ICCBR 1997. Lecture Notes in Computer Science, vol 1266. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63233-6_514
Download citation
DOI: https://doi.org/10.1007/3-540-63233-6_514
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63233-7
Online ISBN: 978-3-540-69238-6
eBook Packages: Springer Book Archive