Online Learning of Multiple Tasks and Their Relationships

Avishek Saha, Piyush Rai, Hal Daumé III, Suresh Venkatasubramanian
Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, PMLR 15:643-651, 2011.

Abstract

We propose an Online MultiTask Learning (OMTL) framework which simultaneously learns the task weight vectors as well as the task relatedness adaptively from the data. Our work is in contrast with prior work on online multitask learning which assumes fixed task relatedness, a priori. Furthermore, whereas prior work in such settings assume only positively correlated tasks, our framework can capture negative correlations as well. Our proposed framework learns the task relationship matrix by framing the objective function as a Bregman divergence minimization problem for positive definite matrices. Subsequently, we exploit this adaptively learned task-relationship matrix to select the most informative samples in an online multitask active learning setting. Experimental results on a number of real-world datasets and comparisons with numerous baselines establish the efficacy of our proposed approach.

Cite this Paper

BibTeX
@InProceedings{pmlr-v15-saha11b, title = {Online Learning of Multiple Tasks and Their Relationships}, author = {Saha, Avishek and Rai, Piyush and III, Hal Daumé and Venkatasubramanian, Suresh}, booktitle = {Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics}, pages = {643--651}, year = {2011}, editor = {Gordon, Geoffrey and Dunson, David and Dudík, Miroslav}, volume = {15}, series = {Proceedings of Machine Learning Research}, address = {Fort Lauderdale, FL, USA}, month = {11--13 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v15/saha11b/saha11b.pdf}, url = {https://proceedings.mlr.press/v15/saha11b.html}, abstract = {We propose an Online MultiTask Learning (OMTL) framework which simultaneously learns the task weight vectors as well as the task relatedness adaptively from the data. Our work is in contrast with prior work on online multitask learning which assumes fixed task relatedness, a priori. Furthermore, whereas prior work in such settings assume only positively correlated tasks, our framework can capture negative correlations as well. Our proposed framework learns the task relationship matrix by framing the objective function as a Bregman divergence minimization problem for positive definite matrices. Subsequently, we exploit this adaptively learned task-relationship matrix to select the most informative samples in an online multitask active learning setting. Experimental results on a number of real-world datasets and comparisons with numerous baselines establish the efficacy of our proposed approach.} }
Endnote
%0 Conference Paper %T Online Learning of Multiple Tasks and Their Relationships %A Avishek Saha %A Piyush Rai %A Hal Daumé III %A Suresh Venkatasubramanian %B Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2011 %E Geoffrey Gordon %E David Dunson %E Miroslav Dudík %F pmlr-v15-saha11b %I PMLR %P 643--651 %U https://proceedings.mlr.press/v15/saha11b.html %V 15 %X We propose an Online MultiTask Learning (OMTL) framework which simultaneously learns the task weight vectors as well as the task relatedness adaptively from the data. Our work is in contrast with prior work on online multitask learning which assumes fixed task relatedness, a priori. Furthermore, whereas prior work in such settings assume only positively correlated tasks, our framework can capture negative correlations as well. Our proposed framework learns the task relationship matrix by framing the objective function as a Bregman divergence minimization problem for positive definite matrices. Subsequently, we exploit this adaptively learned task-relationship matrix to select the most informative samples in an online multitask active learning setting. Experimental results on a number of real-world datasets and comparisons with numerous baselines establish the efficacy of our proposed approach.
RIS
TY - CPAPER TI - Online Learning of Multiple Tasks and Their Relationships AU - Avishek Saha AU - Piyush Rai AU - Hal Daumé III AU - Suresh Venkatasubramanian BT - Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics DA - 2011/06/14 ED - Geoffrey Gordon ED - David Dunson ED - Miroslav Dudík ID - pmlr-v15-saha11b PB - PMLR DP - Proceedings of Machine Learning Research VL - 15 SP - 643 EP - 651 L1 - http://proceedings.mlr.press/v15/saha11b/saha11b.pdf UR - https://proceedings.mlr.press/v15/saha11b.html AB - We propose an Online MultiTask Learning (OMTL) framework which simultaneously learns the task weight vectors as well as the task relatedness adaptively from the data. Our work is in contrast with prior work on online multitask learning which assumes fixed task relatedness, a priori. Furthermore, whereas prior work in such settings assume only positively correlated tasks, our framework can capture negative correlations as well. Our proposed framework learns the task relationship matrix by framing the objective function as a Bregman divergence minimization problem for positive definite matrices. Subsequently, we exploit this adaptively learned task-relationship matrix to select the most informative samples in an online multitask active learning setting. Experimental results on a number of real-world datasets and comparisons with numerous baselines establish the efficacy of our proposed approach. ER -
APA
Saha, A., Rai, P., III, H.D. & Venkatasubramanian, S.. (2011). Online Learning of Multiple Tasks and Their Relationships. Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 15:643-651 Available from https://proceedings.mlr.press/v15/saha11b.html.

Related Material