Effective solutions for real-world stackelberg games: When agents must deal with human uncertainties

James Pita, Manish Jain, Fernando Ordóñez, Milind Tambe, Sarit Kraus, Reuma Magori-Cohen

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

61 Scopus citations

Abstract

How do we build multiagent algorithms for agent interactions with human adversaries? Slackelberg games are natural models for many important applications that involve human interaction, such as oligopolistic markets and security domains. In Stackelberg games, one player, the leader, commits to a strategy and the follower makes their decision with knowledge of the leader's commitment. Existing algorithms for Stackelberg games efficiently find optimal solutions (leader strategy), but they critically assume that the follower plays optimally. Unfortunately, in real-world applications, agents face human followers (adversaries) who - because of their bounded rationality and limited observation of the leader strategy - may deviate from their expected optimal response. Not taking into account these likely deviations when dealing with human adversaries can cause an unacceptable degradation in the leader's reward, particularly in security applications where these algorithms have seen real-world deployment. To address this crucial problem, this paper introduces three new mixed-integer linear programs (MILPs) for Stackelberg games to consider human adversaries, incorporating: (i) novel anchoring theories on human perception of probability distributions and (ii) robustness approaches for MILPs to address human imprecision. Since these new approaches consider human adversaries, traditional proofs of correctness or optimally are insufficient; instead, it is necessary to rely on empirical validation. To that end, this paper considers two settings based on real deployed security systems, and compares 6 different approaches (three new with three previous approaches), in 4 different observability conditions, involving 98 human subjects playing 1360 games in total. The final conclusion was that a model which incorporates both the ideas of robustness and anchoring achieves statistically significant better rewards and also maintains equivalent or faster solution speeds compared to existing approaches.

Original languageEnglish
Title of host publication8th International Joint Conference on Autonomous Agents and Multiagent Systems 2009, AAMAS 2009
PublisherInternational Foundation for Autonomous Agents and Multiagent Systems (IFAAMAS)
Pages515-522
Number of pages8
ISBN (Print)9781615673346
StatePublished - 2009
Externally publishedYes
Event8th International Joint Conference on Autonomous Agents and Multiagent Systems 2009, AAMAS 2009 - Budapest, Hungary
Duration: 10 May 200915 May 2009

Publication series

NameProceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS
Volume1
ISSN (Print)1548-8403
ISSN (Electronic)1558-2914

Conference

Conference8th International Joint Conference on Autonomous Agents and Multiagent Systems 2009, AAMAS 2009
Country/TerritoryHungary
CityBudapest
Period10/05/0915/05/09

Keywords

  • Bayesian and stackel- berg games
  • Game theory
  • Security of agent systems

Fingerprint

Dive into the research topics of 'Effective solutions for real-world stackelberg games: When agents must deal with human uncertainties'. Together they form a unique fingerprint.

Cite this