University of Sussex
Browse
Quaetal11.pdf (196.08 kB)

Multitask learning without label correspondences

Download (196.08 kB)
conference contribution
posted on 2023-06-08, 16:45 authored by Novi QuadriantoNovi Quadrianto, Alexander Smola, Tiberio Caetano, S V N Vishwanathan, James Petterson
We propose an algorithm to perform multitask learning where each task has potentially distinct label sets and label correspondences are not readily available. This is in contrast with existing methods which either assume that the label sets shared by different tasks are the same or that there exists a label mapping oracle. Our method directly maximizes the mutual information among the labels, and we show that the resulting objective function can be efficiently optimized using existing algorithms. Our proposed approach has a direct application for data integration with different label spaces for the purpose of classification, such as integrating Yahoo! and DMOZ web directories.

History

Publication status

  • Published

Journal

Proceedings of the Advances in Neural Information Processing Systems 23; Vancouver, British Columbia, Canada; 6-9 December 2010

Publisher

Neural Information Processing Systems Foundation

Issue

23

Volume

1

Page range

1957-1965

Pages

2631.0

Book title

Proceedings of the 24th annual conference on neural information processing systems 2010

Place of publication

Red Hook, NY

ISBN

9781617823800

Series

Advances in neural information processing systems

Department affiliated with

  • Informatics Publications

Full text available

  • Yes

Peer reviewed?

  • Yes

Editors

Chris Williams, Aron Culotta, John Shawe-Taylor, John Lafferty, Richard Zemel

Legacy Posted Date

2014-02-24

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC