Learning under persistent drift
In this paper we study learning algorithms for environments which are changing over time. Unlike most previous work, we are interested in the case where the changes might be rapid but their “direction” is relatively constant. We model this type of change by assuming that the target distribution is changing continuously at a constant rate from one extreme distribution to another. We show in this case how to use a simple weighting scheme to estimate the error of an hypothesis, and using this estimate, to minimize the error of the prediction.
Unable to display preview. Download preview PDF.
- [Bar92]Bartlett. Learning with slowly changing distribution. In Proceedings of the Workshop on Computational Learning Theory, Morgan Kaufmann Publishers, pages 243–252, 1992.Google Scholar
- [BBDK96]Peter Bartlett, Shai Ben-David, and Sanjeev Kulkarni. Learning changing concepts by exploiting the structure of change. In Proceedings of the Workshop on Computational Learning Theory, Morgan Kaufmann Publishers, 1996.Google Scholar
- [BL96]Rakesh D. Barve and Philip M. Long. On the complexity of learning from drifting distributions. In Proceedings of the Workshop on Computational Learning Theory, Morgan Kaufmann Publishers, 1996.Google Scholar
- [Hau92]David Haussler. Decision theoretic generalization of the pac model for neural net and other learning applications. Information and Computation, 100:78–150, 1992.Google Scholar
- [HL94]David Helmbold and Phill Long. Tracking drifting concepts by minimizing disagreements. Machine Learning, 14(1):27–46, 1994. A preliminary version appeared in Proceedings of COLT 1991, 13–23.Google Scholar
- [Pol84]David Pollard. Convergence of Stochastic Processes. Springer-Verlag, 1984.Google Scholar
- [Val84]Leslie G. Valiant. A theory of the learnable. Communications of the ACM, 27(11):1134–1142, November 1984.Google Scholar
- [Vap82]V. N. Vapnik. Estimation of Dependences Based on Empirical Data. Springer-Verlag, New York, 1982.Google Scholar