Learning coordinate gradients with multi-task kernels

Ying Yiming, ICG Campbell

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

6 Citations (Scopus)

Abstract

Coordinate gradient learning is motivated by the problem of variable selection and determining variable covariation. In this paper we propose a novel unifying framework for coordinate gradient learning (MGL) from the perspective of multi-task learning. Our approach relies on multi-task kernels to simulate the structure of gradient learning. This has several appealing properties. Firstly, it allows us to introduce a novel algorithm which appropriately captures the inherent structure of coordinate gradient learning. Secondly, this approach gives rise to a clear algorithmic process: a computational optimization algorithm which is memory and time efficient. Finally, a statistical error analysis ensures convergence of the estimated function and its gradient to the true function and true gradient. We report some preliminary experiments to validate MGL for variable selection as well as determining variable covariation.
Translated title of the contributionLearning coordinate gradients with multi-task kernels
Original languageEnglish
Title of host publicationCOLT 2008
Pages217 - 228
Number of pages11
Publication statusPublished - 2008

Bibliographical note

Name and Venue of Event: COLT
Conference Proceedings/Title of Journal: Proceedings: Computational Learning Theory (COLT)

Fingerprint Dive into the research topics of 'Learning coordinate gradients with multi-task kernels'. Together they form a unique fingerprint.

  • Cite this