Automated Planning in Repeated Adversarial Games

Enrique Munoz de Cote, Adam M. Sykulski, Archie Chapman, Nick Jennings
Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence, PMLR R8:151-158, 2010.

Abstract

Game theory’s prescriptive power typically relies on full rationality and/or self–play interactions. In contrast, this work sets aside these fundamental premises and focuses instead on heterogeneous autonomous interactions between two or more agents. Specifically, we introduce a new and concise representation for repeated adversarial (constant–sum) games that highlight the necessary features that enable an automated planing agent to reason about how to score above the game’s Nash equilibrium, when facing heterogeneous adversaries. To this end, we present TeamUP, a model–based RL algorithm designed for learning and planning such an abstraction. In essence, it is somewhat similar to R-max with a cleverly engineered reward shaping that treats exploration as an adversarial optimization problem. In practice, it attempts to find an ally with which to tacitly collude (in more than two–player games) and then collaborates on a joint plan of actions that can consistently score a high utility in adversarial repeated games. We use the inaugural Lemonade Stand Game Tournament1 to demonstrate the effectiveness of our approach, and find that TeamUP is the best performing agent, demoting the Tournament’s actual winning strategy into second place. In our experimental analysis, we show hat our strategy successfully and consistently builds collaborations with many different heterogeneous (and sometimes very sophisticated) adversaries.

Cite this Paper

BibTeX
@InProceedings{pmlr-vR8-cote10a, title = {Automated Planning in Repeated Adversarial Games}, author = {de Cote, Enrique Munoz and Sykulski, Adam M. and Chapman, Archie and Jennings, Nick}, booktitle = {Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence}, pages = {151--158}, year = {2010}, editor = {Grünwald, Peter and Spirtes, Peter}, volume = {R8}, series = {Proceedings of Machine Learning Research}, month = {08--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/r8/main/assets/cote10a/cote10a.pdf}, url = {https://proceedings.mlr.press/r8/cote10a.html}, abstract = {Game theory’s prescriptive power typically relies on full rationality and/or self–play interactions. In contrast, this work sets aside these fundamental premises and focuses instead on heterogeneous autonomous interactions between two or more agents. Specifically, we introduce a new and concise representation for repeated adversarial (constant–sum) games that highlight the necessary features that enable an automated planing agent to reason about how to score above the game’s Nash equilibrium, when facing heterogeneous adversaries. To this end, we present TeamUP, a model–based RL algorithm designed for learning and planning such an abstraction. In essence, it is somewhat similar to R-max with a cleverly engineered reward shaping that treats exploration as an adversarial optimization problem. In practice, it attempts to find an ally with which to tacitly collude (in more than two–player games) and then collaborates on a joint plan of actions that can consistently score a high utility in adversarial repeated games. We use the inaugural Lemonade Stand Game Tournament1 to demonstrate the effectiveness of our approach, and find that TeamUP is the best performing agent, demoting the Tournament’s actual winning strategy into second place. In our experimental analysis, we show hat our strategy successfully and consistently builds collaborations with many different heterogeneous (and sometimes very sophisticated) adversaries.}, note = {Reissued by PMLR on 04 October 2026.} }
Endnote
%0 Conference Paper %T Automated Planning in Repeated Adversarial Games %A Enrique Munoz de Cote %A Adam M. Sykulski %A Archie Chapman %A Nick Jennings %B Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2010 %E Peter Grünwald %E Peter Spirtes %F pmlr-vR8-cote10a %I PMLR %P 151--158 %U https://proceedings.mlr.press/r8/cote10a.html %V R8 %X Game theory’s prescriptive power typically relies on full rationality and/or self–play interactions. In contrast, this work sets aside these fundamental premises and focuses instead on heterogeneous autonomous interactions between two or more agents. Specifically, we introduce a new and concise representation for repeated adversarial (constant–sum) games that highlight the necessary features that enable an automated planing agent to reason about how to score above the game’s Nash equilibrium, when facing heterogeneous adversaries. To this end, we present TeamUP, a model–based RL algorithm designed for learning and planning such an abstraction. In essence, it is somewhat similar to R-max with a cleverly engineered reward shaping that treats exploration as an adversarial optimization problem. In practice, it attempts to find an ally with which to tacitly collude (in more than two–player games) and then collaborates on a joint plan of actions that can consistently score a high utility in adversarial repeated games. We use the inaugural Lemonade Stand Game Tournament1 to demonstrate the effectiveness of our approach, and find that TeamUP is the best performing agent, demoting the Tournament’s actual winning strategy into second place. In our experimental analysis, we show hat our strategy successfully and consistently builds collaborations with many different heterogeneous (and sometimes very sophisticated) adversaries. %Z Reissued by PMLR on 04 October 2026.
APA
de Cote, E.M., Sykulski, A.M., Chapman, A. & Jennings, N.. (2010). Automated Planning in Repeated Adversarial Games. Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research R8:151-158 Available from https://proceedings.mlr.press/r8/cote10a.html. Reissued by PMLR on 04 October 2026.

Related Material