Multitask classification by learning the task relevance

Jun Fang, Shihao Ji, Ya Xue, Lawrence Carin

Research output: Contribution to journalArticlepeer-review

6 Scopus citations

Abstract

We consider the problem of multitask learning (MTL), in which we simultaneously learn classifiers for multiple data sets (tasks), with sharing of intertask data as appropriate. We introduce a set of relevance parameters that control the degree to which data from other tasks are used in estimating the current task's classifier parameters. The set of relevance parameters are learned by maximizing their posterior probability, yielding an expectation-maximization (EM) algorithm. We illustrate the effectiveness of our approach through experimental results on a practical data set. © 2008 IEEE.
Original languageEnglish (US)
Pages (from-to)593-596
Number of pages4
JournalIEEE Signal Processing Letters
Volume15
DOIs
StatePublished - Dec 1 2008
Externally publishedYes

Fingerprint Dive into the research topics of 'Multitask classification by learning the task relevance'. Together they form a unique fingerprint.

Cite this