Multitask Learning Without Label Correspondences
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.
Cite
Text
Quadrianto et al. "Multitask Learning Without Label Correspondences." Neural Information Processing Systems, 2010.Markdown
[Quadrianto et al. "Multitask Learning Without Label Correspondences." Neural Information Processing Systems, 2010.](https://mlanthology.org/neurips/2010/quadrianto2010neurips-multitask/)BibTeX
@inproceedings{quadrianto2010neurips-multitask,
title = {{Multitask Learning Without Label Correspondences}},
author = {Quadrianto, Novi and Petterson, James and Caetano, Tibério S. and Smola, Alex J. and Vishwanathan, S.v.n.},
booktitle = {Neural Information Processing Systems},
year = {2010},
pages = {1957-1965},
url = {https://mlanthology.org/neurips/2010/quadrianto2010neurips-multitask/}
}