Abstract
We introduce a framework for class noise, in which most of the known class noise models for the PAC setting can be formulated. Within this framework, we study properties of noise models that enable learning of concept classes of finite VC-dimension with the Empirical Risk Minimization (ERM) strategy. We introduce simple noise models for which classical ERM is not successful. Aiming at a more generalpurpose algorithm for learning under noise, we generalize ERM to a more powerful strategy. Finally, we study general characteristics of noise models that enable learning of concept classes of finite VC-dimension with this new strategy.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Angluin, D., Laird, P.: Learning from noisy examples. Machine Learning 2, 343–370 (1988)
Ben-David, S., Shalev-Shwartz, S., Urner, R.: Domain adaptation–can quantity compensate for quality? In: ISAIM (2012)
Blumer, A., Ehrenfeucht, A., Haussler, D., Warmuth, M.: Classifying learnable geometric concepts with the Vapnik-Chervonenkis dimension. In: STOC, pp. 273–282 (1986)
Bshouty, N., Eiron, N., Kushilevitz, E.: PAC learning with nasty noise. Theoretical Computer Science 288, 255–275 (2002)
Crammer, K., Kearns, M., Wortman, J.: Learning from data of variable quality. In: NIPS, pp. 219–226 (2005)
Decatur, S.: PAC learning with constant-partition classification noise and applications to decision tree induction. In: ICML, pp. 83–91 (1997)
Goldman, S., Sloan, R.: Can PAC learning algorithms tolerate random attribute noise? Algorithmica 14, 70–84 (1995)
Haussler, D.: Decision theoretic generalizations of the PAC model for neural net and other learning applications. Information and Computation 100, 78–150 (1992)
Kearns, M.: Efficient noise-tolerant learning from statistical queries. Journal of the ACM 45, 983–1006 (1998)
Kearns, M., Schapire, R.: Efficient distribution-free learning of probabilistic concepts. In: SFCS, pp. 382–391 (1990)
Kearns, M., Schapire, R., Sellie, L.: Toward efficient agnostic learning. Machine Learning 17, 115–141 (1994)
Laird, P.: Learning from Good and Bad Data. Kluwer Academic Publishers (1988)
Ralaivola, L., Denis, F., Magnan, C.: CN = CPCN. In: ICML, pp. 721–728 (2006)
Sloan, R.: Four types of noise in data for PAC learning. Information Processing Letters 54, 157–162 (1995)
Tsybakov, A.: Optimal aggregation of classifiers in statistical learning. Annals of Statistics 32, 135–166 (2004)
Valiant, L.: A theory of the learnable. In: STOC, pp. 436–445 (1984)
Vapnik, V.: The Nature of Statistical Learning Theory. Springer (1995)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2012 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Jabbari, S., Holte, R.C., Zilles, S. (2012). PAC-Learning with General Class Noise Models. In: Glimm, B., Krüger, A. (eds) KI 2012: Advances in Artificial Intelligence. KI 2012. Lecture Notes in Computer Science(), vol 7526. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-33347-7_7
Download citation
DOI: https://doi.org/10.1007/978-3-642-33347-7_7
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-33346-0
Online ISBN: 978-3-642-33347-7
eBook Packages: Computer ScienceComputer Science (R0)