Abstract
We describe a framework for solving regression problems by reduction to classification. Our reduction is based on symmetrization of margin-based loss functions commonly used in boosting algorithms, namely, the logistic loss and the exponential loss. Our construction yields a smooth version of the ε-insensitive hinge loss that is used in support vector regression. A byproduct of this construction is a new simple form of regularization for boosting-based classification and regression algorithms. We present two parametric families of batch learning algorithms for minimizing these losses. The first family employs a log-additive update and is based on recent boosting algorithms while the second family uses a new form of additive update. We also describe and analyze online gradient descent (GD) and exponentiated gradient (EG) algorithms for the ε-insensitive logistic loss. Our regression framework also has implications on classification algorithms, namely, a new additive batch algorithm for the log-loss and exp-loss used in boosting.
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
Cesa-Bianchi, N.: Analysis of two gradient-based algorithms for on-line regression. Journal of Computer and System Sciences 59(3), 392–411 (1999)
Collins, M., Schapire, R.E., Singer, Y.: Logistic regression, AdaBoost and Bregman distances. Machine Learning 47(2/3), 253–285 (2002)
Duffy, N., Helmbold, D.: Leveraging for regression. In: Proceedings of the 15th Annual Conference on Computational Learning Theory. ACM, New York (2000)
Freund, Y., Opper, M.: Drifting games and Brownian motion. Journal of Computer and System Sciences 64, 113–132 (2002)
Friedman, J., Hastie, T., Tibshirani, R.: Additive logistic regression: a statistical view of boosting. Annals of Statistics 28(2), 337–374 (2000)
Hastie, T., Tibshirani, R., Friedman, J.: The Elements of Statistical Learning. Springer, Heidelberg (2001)
Huber, P.J.: Robust Statistics. John Wiley and Sons, New York (1981)
Kivinen, J., Helmbold, D.P., Warmuth, M.: P Helmbold, and M. Warmuth. Relative loss bounds for single neurons. IEEE Transactions on Neural Networks 10(6), 1291–1304 (1999)
Kivinen, J., Warmuth, M.K.: Exponentiated gradient versus gradient descent for linear predictors. Information and Computation 132(1), 1–64 (1997)
Lebanon, G., Lafferty, J.: Boosting and maximum likelihood for exponential models. In: Advances in Neural Information Processing Systems, vol. 14 (2001)
Poggio, T., Girosi, F.: Networks for approximation and learning. Proceedings of the IEEE 78(9) (1990)
Robert, E.: Schapire. Drifting games. In: Proceedings of the 12th Annual Conference on Computational Learning Theory (1999)
Smola, A., Schölkopf, B.: A tutorial on support vector regression. Technical Report NC2-TR-1998-030, NeuroCOLT2 (1998)
Vapnik, V.N.: The Nature of Statistical Learning Theory. Springer, Heidelberg (1995)
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
Dekel, O., Shalev-Shwartz, S., Singer, Y. (2003). Smooth ε-Insensitive Regression by Loss Symmetrization. In: Schölkopf, B., Warmuth, M.K. (eds) Learning Theory and Kernel Machines. Lecture Notes in Computer Science(), vol 2777. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-45167-9_32
Download citation
DOI: https://doi.org/10.1007/978-3-540-45167-9_32
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-40720-1
Online ISBN: 978-3-540-45167-9
eBook Packages: Springer Book Archive