Bayesian Multitask Learning with Latent Hierarchies

Hal Daume
Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence, PMLR R7:135-142, 2009.

Abstract

We learn multiple hypotheses for related tasks under a latent hierarchical relationship between tasks. We exploit the intuition that for domain adaptation, we wish to share classifier structure, but for multitask learning, we wish to share covariance structure. Our hierarchical model is seen to subsume several previously proposed multitask learning models and performs well on three distinct real-world data sets.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR7-daume09a, title = {{B}ayesian Multitask Learning with Latent Hierarchies}, author = {Daume, Hal}, booktitle = {Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence}, pages = {135--142}, year = {2009}, editor = {Bilmes, Jeff and Ng, Andrew Y.}, volume = {R7}, series = {Proceedings of Machine Learning Research}, month = {18--21 Jun}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r7/main/assets/daume09a/daume09a.pdf}, url = {https://proceedings.mlr.press/r7/daume09a.html}, abstract = {We learn multiple hypotheses for related tasks under a latent hierarchical relationship between tasks. We exploit the intuition that for domain adaptation, we wish to share classifier structure, but for multitask learning, we wish to share covariance structure. Our hierarchical model is seen to subsume several previously proposed multitask learning models and performs well on three distinct real-world data sets.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Bayesian Multitask Learning with Latent Hierarchies %A Hal Daume %B Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2009 %E Jeff Bilmes %E Andrew Y. Ng %F pmlr-vR7-daume09a %I PMLR %P 135--142 %U https://proceedings.mlr.press/r7/daume09a.html %V R7 %X We learn multiple hypotheses for related tasks under a latent hierarchical relationship between tasks. We exploit the intuition that for domain adaptation, we wish to share classifier structure, but for multitask learning, we wish to share covariance structure. Our hierarchical model is seen to subsume several previously proposed multitask learning models and performs well on three distinct real-world data sets. %Z Reissued by PMLR on 04 October 2026.
APA
Daume, H.. (2009). Bayesian Multitask Learning with Latent Hierarchies. Proceedings of the 25th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R7:135-142 Available from https://proceedings.mlr.press/r7/daume09a.html. Reissued by PMLR on 04 October 2026.

Related Material