PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

A case study on Meta-Generalizing: a Gaussian Processes approach
Grigorios Skolidis and Guido Sanguinetti
Journal of Machine learning research 2012.

Abstract

We propose a novel model for meta-generalisation, i.e. performing prediction on novel tasks based on information from multiple different but related tasks. The model is based on two coupled Gaussian processes with structured covariance function; one model performs predictions by learning a constrained covariance function encapsulating the relations between the various training tasks, while the second model determines the similarity of new tasks to previously seen tasks. We demonstrate empirically on several real and synthetic data sets both the strengths of the approach and its limitations due to the distributional assumptions underpinning it.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:8969
Deposited By:Grigorios Skolidis
Deposited On:21 February 2012