PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

New Adaptive Algorithms for Online Classification
Francesco Orabona and Koby Crammer
In: Neural Information Processing Systems, 13-15 Dec 2011, Vancouver, Canada.


We propose a general framework to online learning for classification problems with time-varying potential functions in the adversarial setting. This framework allows to design and prove relative mistake bounds for any generic loss function. The mistake bounds can be specialized for the hinge loss, allowing to recover and improve the bounds of known online classification algorithms. By optimizing the general bound we derive a new online classification algorithm, called NAROW, that hybridly uses adaptive- and fixed- second order information. We analyze the properties of the algorithm and illustrate its performance using synthetic dataset.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Poster)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Natural Language Processing
Theory & Algorithms
ID Code:7695
Deposited By:Francesco Orabona
Deposited On:17 March 2011