PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Stochastic Low-Rank Kernel Learning for Regression
Pierre Machart, Thomas Peel, Sandrine Anthoine, Liva Ralaivola and Hervé Glotin
In: ICML 2011, 28 Jun - 02 Jul 2011, Bellevue, Washington.


We present a novel approach to learn a kernel-based regression function. It is based on the use of conical combinations of data-based parameterized kernels and on a new stochastic convex optimization procedure of which we establish convergence guarantees. The overall learning procedure has the nice properties that a) the learned conical combination is automatically designed to perform the regression task at hand and b) the updates implicated by the optimization procedure are quite inexpensive. In order to shed light on the appositeness of our learning strategy, we present empirical results from experiments conducted on various benchmark datasets.

EPrint Type:Conference or Workshop Item (Lecture)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:8765
Deposited By:Pierre Machart
Deposited On:21 February 2012