A near-optimal algorithm for finite partial-monitoring games against adversarial opponents

Gábor Bartók
Proceedings of the 26th Annual Conference on Learning Theory, PMLR 30:696-710, 2013.

Abstract

Partial monitoring is an online learning model where in every time step, after a learner and an opponent choose their actions, the loss and the feedback for the learner is calculated based on a loss and a feedback function, both of which are known to the learner ahead of time. As in other online learning scenarios, the goal of the learner is to minimize his cumulative loss. In this paper we present and analyze a new algorithm for locally observable partial monitoring games. We prove that the expected regret of our algorithm is of \tilde O(\sqrtN’T), where T is the time horizon and N’ is the size of the largest point-local game. The most important improvement of this bound compared to previous results is that it does not depend directly on the number of actions, but rather on the structure of the game.

Cite this Paper


BibTeX
@InProceedings{pmlr-v30-Bartok13, title = {A near-optimal algorithm for finite partial-monitoring games against adversarial opponents}, author = {Bartók, Gábor}, booktitle = {Proceedings of the 26th Annual Conference on Learning Theory}, pages = {696--710}, year = {2013}, editor = {Shalev-Shwartz, Shai and Steinwart, Ingo}, volume = {30}, series = {Proceedings of Machine Learning Research}, address = {Princeton, NJ, USA}, month = {12--14 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v30/Bartok13.pdf}, url = {https://proceedings.mlr.press/v30/Bartok13.html}, abstract = {Partial monitoring is an online learning model where in every time step, after a learner and an opponent choose their actions, the loss and the feedback for the learner is calculated based on a loss and a feedback function, both of which are known to the learner ahead of time. As in other online learning scenarios, the goal of the learner is to minimize his cumulative loss. In this paper we present and analyze a new algorithm for locally observable partial monitoring games. We prove that the expected regret of our algorithm is of \tilde O(\sqrtN’T), where T is the time horizon and N’ is the size of the largest point-local game. The most important improvement of this bound compared to previous results is that it does not depend directly on the number of actions, but rather on the structure of the game.} }
Endnote
%0 Conference Paper %T A near-optimal algorithm for finite partial-monitoring games against adversarial opponents %A Gábor Bartók %B Proceedings of the 26th Annual Conference on Learning Theory %C Proceedings of Machine Learning Research %D 2013 %E Shai Shalev-Shwartz %E Ingo Steinwart %F pmlr-v30-Bartok13 %I PMLR %P 696--710 %U https://proceedings.mlr.press/v30/Bartok13.html %V 30 %X Partial monitoring is an online learning model where in every time step, after a learner and an opponent choose their actions, the loss and the feedback for the learner is calculated based on a loss and a feedback function, both of which are known to the learner ahead of time. As in other online learning scenarios, the goal of the learner is to minimize his cumulative loss. In this paper we present and analyze a new algorithm for locally observable partial monitoring games. We prove that the expected regret of our algorithm is of \tilde O(\sqrtN’T), where T is the time horizon and N’ is the size of the largest point-local game. The most important improvement of this bound compared to previous results is that it does not depend directly on the number of actions, but rather on the structure of the game.
RIS
TY - CPAPER TI - A near-optimal algorithm for finite partial-monitoring games against adversarial opponents AU - Gábor Bartók BT - Proceedings of the 26th Annual Conference on Learning Theory DA - 2013/06/13 ED - Shai Shalev-Shwartz ED - Ingo Steinwart ID - pmlr-v30-Bartok13 PB - PMLR DP - Proceedings of Machine Learning Research VL - 30 SP - 696 EP - 710 L1 - http://proceedings.mlr.press/v30/Bartok13.pdf UR - https://proceedings.mlr.press/v30/Bartok13.html AB - Partial monitoring is an online learning model where in every time step, after a learner and an opponent choose their actions, the loss and the feedback for the learner is calculated based on a loss and a feedback function, both of which are known to the learner ahead of time. As in other online learning scenarios, the goal of the learner is to minimize his cumulative loss. In this paper we present and analyze a new algorithm for locally observable partial monitoring games. We prove that the expected regret of our algorithm is of \tilde O(\sqrtN’T), where T is the time horizon and N’ is the size of the largest point-local game. The most important improvement of this bound compared to previous results is that it does not depend directly on the number of actions, but rather on the structure of the game. ER -
APA
Bartók, G.. (2013). A near-optimal algorithm for finite partial-monitoring games against adversarial opponents. Proceedings of the 26th Annual Conference on Learning Theory, in Proceedings of Machine Learning Research 30:696-710 Available from https://proceedings.mlr.press/v30/Bartok13.html.

Related Material