Multitask classification by learning the task relevance
Publication
, Journal Article
Fang, J; Ji, S; Xue, Y; Carin, L
Published in: IEEE Signal Processing Letters
December 1, 2008
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.
Duke Scholars
Published In
IEEE Signal Processing Letters
DOI
ISSN
1070-9908
Publication Date
December 1, 2008
Volume
15
Start / End Page
593 / 596
Related Subject Headings
- Networking & Telecommunications
- 4603 Computer vision and multimedia computation
- 4009 Electronics, sensors and digital hardware
- 4006 Communications engineering
- 1005 Communications Technologies
- 0906 Electrical and Electronic Engineering
- 0801 Artificial Intelligence and Image Processing
Citation
APA
Chicago
ICMJE
MLA
NLM
Fang, J., Ji, S., Xue, Y., & Carin, L. (2008). Multitask classification by learning the task relevance. IEEE Signal Processing Letters, 15, 593–596. https://doi.org/10.1109/LSP.2008.2001967
Fang, J., S. Ji, Y. Xue, and L. Carin. “Multitask classification by learning the task relevance.” IEEE Signal Processing Letters 15 (December 1, 2008): 593–96. https://doi.org/10.1109/LSP.2008.2001967.
Fang J, Ji S, Xue Y, Carin L. Multitask classification by learning the task relevance. IEEE Signal Processing Letters. 2008 Dec 1;15:593–6.
Fang, J., et al. “Multitask classification by learning the task relevance.” IEEE Signal Processing Letters, vol. 15, Dec. 2008, pp. 593–96. Scopus, doi:10.1109/LSP.2008.2001967.
Fang J, Ji S, Xue Y, Carin L. Multitask classification by learning the task relevance. IEEE Signal Processing Letters. 2008 Dec 1;15:593–596.
Published In
IEEE Signal Processing Letters
DOI
ISSN
1070-9908
Publication Date
December 1, 2008
Volume
15
Start / End Page
593 / 596
Related Subject Headings
- Networking & Telecommunications
- 4603 Computer vision and multimedia computation
- 4009 Electronics, sensors and digital hardware
- 4006 Communications engineering
- 1005 Communications Technologies
- 0906 Electrical and Electronic Engineering
- 0801 Artificial Intelligence and Image Processing