Active Learning for Developing Personalized Treatment

Kun Deng, Joelle Pineau, Susan A. Murphy
Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence, PMLR R9:189-196, 2011.

Abstract

The personalization of treatment via bio-markers and other risk categories has drawn increasing interest among clinical scientists. Personalized treatment strategies can be learned using data from clinical trials, but such trials are very costly to run. This paper explores the use of active learning techniques to design more efficient trials, addressing issues such as whom to recruit, at what point in the trial, and which treatment to assign, throughout the duration of the trial. We propose a minimax bandit model with two different optimization criteria, and discuss the computational challenges and issues pertaining to this approach. We evaluate our active learning policies using both simulated data, and data modeled after a clinical trial for treating depressed individuals, and contrast our methods with other plausible active learning policies.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR9-deng11a, title = {Active Learning for Developing Personalized Treatment}, author = {Deng, Kun and Pineau, Joelle and Murphy, Susan A.}, booktitle = {Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence}, pages = {189--196}, year = {2011}, editor = {Cozman, Fabio and Pfeffer, Avi}, volume = {R9}, series = {Proceedings of Machine Learning Research}, month = {14--17 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r9/main/assets/deng11a/deng11a.pdf}, url = {https://proceedings.mlr.press/r9/deng11a.html}, abstract = {The personalization of treatment via bio-markers and other risk categories has drawn increasing interest among clinical scientists. Personalized treatment strategies can be learned using data from clinical trials, but such trials are very costly to run. This paper explores the use of active learning techniques to design more efficient trials, addressing issues such as whom to recruit, at what point in the trial, and which treatment to assign, throughout the duration of the trial. We propose a minimax bandit model with two different optimization criteria, and discuss the computational challenges and issues pertaining to this approach. We evaluate our active learning policies using both simulated data, and data modeled after a clinical trial for treating depressed individuals, and contrast our methods with other plausible active learning policies.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Active Learning for Developing Personalized Treatment %A Kun Deng %A Joelle Pineau %A Susan A. Murphy %B Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2011 %E Fabio Cozman %E Avi Pfeffer %F pmlr-vR9-deng11a %I PMLR %P 189--196 %U https://proceedings.mlr.press/r9/deng11a.html %V R9 %X The personalization of treatment via bio-markers and other risk categories has drawn increasing interest among clinical scientists. Personalized treatment strategies can be learned using data from clinical trials, but such trials are very costly to run. This paper explores the use of active learning techniques to design more efficient trials, addressing issues such as whom to recruit, at what point in the trial, and which treatment to assign, throughout the duration of the trial. We propose a minimax bandit model with two different optimization criteria, and discuss the computational challenges and issues pertaining to this approach. We evaluate our active learning policies using both simulated data, and data modeled after a clinical trial for treating depressed individuals, and contrast our methods with other plausible active learning policies. %Z Reissued by PMLR on 04 October 2026.
APA
Deng, K., Pineau, J. & Murphy, S.A.. (2011). Active Learning for Developing Personalized Treatment. Proceedings of the 27th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R9:189-196 Available from https://proceedings.mlr.press/r9/deng11a.html. Reissued by PMLR on 04 October 2026.

Related Material