Multitask learning without label correspondences

Quadrianto, Novi, Smola, Alexander, Caetano, Tiberio, Vishwanathan, S V N and Petterson, James (2011) Multitask learning without label correspondences. Published in: Lafferty, John, Williams, Chris, Shawe-Taylor, John, Zemel, Richard and Culotta, Aron, (eds.) Proceedings of the Advances in Neural Information Processing Systems 23; Vancouver, British Columbia, Canada; 6-9 December 2010. 1 (23) 1957-1965. Neural Information Processing Systems Foundation, Red Hook, NY. ISBN 9781617823800

[img]
Preview
PDF
Download (200kB) | Preview

Abstract

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.

Item Type: Conference Proceedings
Schools and Departments: School of Engineering and Informatics > Informatics
Subjects: Q Science > Q Science (General)
Depositing User: Novi Quadrianto
Date Deposited: 24 Feb 2014 14:09
Last Modified: 16 Jun 2017 15:23
URI: http://sro.sussex.ac.uk/id/eprint/47611

View download statistics for this item

📧 Request an update