We study online learning when individual instances are corrupted by adversarially chosen random noise. We assume the noise distribution is unknown, and may change over time with no restriction other than having zero mean and bounded variance. Our technique relies on a family of unbiased estimators for non-linear functions, which may be of independent interest. We show that a variant of online gradient descent can learn functions in any dotproduct (e.g., polynomial) or Gaussian kernel space with any analytic convex loss function. Our variant uses randomized estimates that need to query a random number of noisy copies of each instance, where with high probability this number is upper bounded by a constant. Allowing such multiple queries cannot be avoided: Indeed, we show that online learning is in general impossible when only one noisy copy of each instance can be accessed.
Online learning of noisy data with kernels / N. Cesa-Bianchi, S. Shalev-Shwartz, O. Shamir - In: COLT 2010 : proceedings of the 23rd annual conference on learning theory, Haifa, Israel, june 27-29, 2010 / [a cura di] A. T. Kalai, M. Mohri. - Madison : Omnipress, 2010. - ISBN 9780982252925. - pp. 218-230 (( Intervento presentato al 23. convegno Annual Conference on Learning Theory tenutosi a Haifa, Israel nel 2010.
|Titolo:||Online learning of noisy data with kernels|
CESA BIANCHI, NICOLO' ANTONIO (Primo)
|Settore Scientifico Disciplinare:||Settore INF/01 - Informatica|
|Data di pubblicazione:||2010|
|Tipologia:||Book Part (author)|
|Appare nelle tipologie:||03 - Contributo in volume|