PROTECT: a deployed game theoretic system to protect the ports of the United States

作者: Ben Maule , Eric Shieh , Craig Baldwin , Garrett Meyer , Joseph DiRenzo

DOI: 10.5555/2343576.2343578

关键词:

摘要: While three deployed applications of game theory for security have recently been reported at AAMAS [12], we as a community remain in the early stages these deployments; there is continuing need to understand core principles innovative theory. Towards that end, this paper presents PROTECT, game-theoretic system by United States Coast Guard (USCG) port Boston scheduling their patrols. USCG has termed deployment PROTECT success, and efforts are underway test it New York, with potential nationwide deployment.PROTECT premised on an attacker-defender Stackelberg model offers five key innovations. First, departure from assumption perfect adversary rationality noted previous work, relying instead quantal response (QR) adversary's behavior --- best our knowledge, first real-world QR model. Second, improve PROTECT's efficiency, generate compact representation defender's strategy space, exploiting equivalence dominance. Third, show how practically real maritime patrolling problem game. Fourth, experimental results illustrate more robustly handles uncertainties than Finally, evaluating time provides data: (i) comparison human-generated vs schedules, (ii) Adversarial Perspective Team's (human mock attackers) analysis.

参考文章(15)
Fernando Ordóñez, Zhengyu Yin, Milind Tambe, Manish Jain, Risk-averse strategies for security games with execution and observational uncertainty national conference on artificial intelligence. pp. 758- 763 ,(2011)
Ondřej Vaněk, Michal Jakob, Ondřej Hrstka, Michal Pěchouček, Using Multi-agent Simulation to Improve the Security of Maritime Transit Multi-Agent-Based Simulation XII. pp. 44- 58 ,(2012) , 10.1007/978-3-642-28400-7_4
Vincent Conitzer, Dmytro Korzhyk, Ronald Parr, Solving Stackelberg games with uncertain observability adaptive agents and multi-agents systems. pp. 1013- 1020 ,(2011) , 10.5555/2034396.2034413
Francesco Amigoni, Nicola Basilico, Nicola Gatti, Leader-follower strategies for robotic patrolling in environments with arbitrary topologies adaptive agents and multi agents systems. ,vol. 1, pp. 57- 64 ,(2009)
Gal A. Kaminka, Noa Agmon, Vladimir Sadov, Sarit Kraus, Adversarial uncertainty in multi-robot patrol international joint conference on artificial intelligence. pp. 1811- 1817 ,(2009)
Christopher Kiekintveld, Richard John, Fernando Ordonez, Rong Yang, Milind Tambe, Improving resource allocation strategy against human adversaries in security games international joint conference on artificial intelligence. pp. 458- 464 ,(2011) , 10.5591/978-1-57735-516-8/IJCAI11-084
Vincent Conitzer, Tuomas Sandholm, Computing the optimal strategy to commit to Proceedings of the 7th ACM conference on Electronic commerce - EC '06. pp. 82- 90 ,(2006) , 10.1145/1134707.1134717