Three Strategies to Success: Learning Adversary Models in Security Games
Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence. Pages 308-313.
Abstract
State-of-the-art applications of Stackelberg security games — including wildlife protection — offer a wealth of data, which can be used to learn the behavior of the adversary. But existing approaches either make strong assumptions about the structure of the data, or gather new data through online algorithms that are likely to play severely suboptimal strategies. We develop a new approach to learning the parameters of the behavioral model of a bounded rational attacker (thereby pinpointing a near optimal strategy), by observing how the attacker responds to only three defender strategies. We also validate our approach using experiments on real and synthetic data.
@inproceedings{ijcai2016p51,
title = {Three Strategies to Success: Learning Adversary Models in Security Games},
author = {Haghtalab, Nika and Fang, Fei and Nguyen, Thanh Hong and Sinha, Arunesh and Procaccia, Ariel D. and Tambe, Milind},
booktitle = {Proceedings of the Twenty-Fifth International Joint Conference on
Artificial Intelligence, {IJCAI-16}},
publisher = {AAAI Press / International Joint Conferences on Artificial Intelligence},
editor = {Subbarao Kambhampati},
pages = {308--313},
year = {2016},
month = {7},
url = {https://ijcai.org/proceedings/2016/51},
}
TY - CPAPER TI - Three Strategies to Success: Learning Adversary Models in Security Games AU - Haghtalab, Nika AU - Fang, Fei AU - Nguyen, Thanh Hong AU - Sinha, Arunesh AU - Procaccia, Ariel D. AU - Tambe, Milind T2 - Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-16 PB - AAAI Press / International Joint Conferences on Artificial Intelligence PY - 2016 SP - 308 EP - 313 UR - https://ijcai.org/proceedings/2016/51 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
