PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Linear algorithms for online multitask classification
Giovanni Cavallanti, Nicolo' Cesa-Bianchi and Claudio Gentile
In: Proc. of the 21st Conference on Learning Theory (COLT'08), Helsinki, Finland(2008).

Abstract

We design and analyze interacting online algorithms for multitask classification that perform better than independent learners whenever the tasks are related in a certain sense. We formalize task relatedness in different ways, and derive formal guarantees on the performance advantage provided by interaction. Our online analysis gives new stimulating insights into previously known co-regularization techniques, such as the multitask kernels and the margin correlation analysis for multiview learning. In the last part we apply our approach to spectral co-regularization: we introduce a natural matrix extension of the quasi-additive algorithm for classification and prove bounds depending on certain unitarily invariant norms of the matrix of task coefficients.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:5215
Deposited By:Claudio Gentile
Deposited On:24 March 2009