TY - GEN
T1 - Coalition Formation in Multi-defender Security Games
AU - Mutzari, Dolev
AU - Gan, Jiarui
AU - Kraus, Sarit
N1 - Publisher Copyright: Copyright © 2021, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
PY - 2021
Y1 - 2021
N2 - We study Stackelberg security game (SSG) with multiple defenders, where heterogeneous defenders need to allocate security resources to protect a set of targets against a strategic attacker. In such games, coordination and cooperation between the defenders can increase their ability to protect their assets, but the heterogeneous preferences of the self-interested defenders often make such cooperation very difficult. In this paper, we approach the problem from the perspective of cooperative game theory and study coalition formation among the defenders. Our main contribution is a number of algorithmic results for the computation problems that arise in this model. We provide a poly-time algorithm for computing a solution in the core of the game and show that all of the elements in the core are Pareto efficient. We show that the problem of computing the entire core is NP-hard and then delve into a special setting where the size of a coalition is limited up to some threshold. We analyse the parameterized complexity of deciding if a coalition structure is in the core under this special setting, and provide a poly-time algorithm for computing successful deviation strategies for a given coalition.
AB - We study Stackelberg security game (SSG) with multiple defenders, where heterogeneous defenders need to allocate security resources to protect a set of targets against a strategic attacker. In such games, coordination and cooperation between the defenders can increase their ability to protect their assets, but the heterogeneous preferences of the self-interested defenders often make such cooperation very difficult. In this paper, we approach the problem from the perspective of cooperative game theory and study coalition formation among the defenders. Our main contribution is a number of algorithmic results for the computation problems that arise in this model. We provide a poly-time algorithm for computing a solution in the core of the game and show that all of the elements in the core are Pareto efficient. We show that the problem of computing the entire core is NP-hard and then delve into a special setting where the size of a coalition is limited up to some threshold. We analyse the parameterized complexity of deciding if a coalition structure is in the core under this special setting, and provide a poly-time algorithm for computing successful deviation strategies for a given coalition.
UR - http://www.scopus.com/inward/record.url?scp=85121733789&partnerID=8YFLogxK
U2 - 10.1609/aaai.v35i6.16704
DO - 10.1609/aaai.v35i6.16704
M3 - منشور من مؤتمر
T3 - 35th AAAI Conference on Artificial Intelligence, AAAI 2021
SP - 5603
EP - 5610
BT - 35th AAAI Conference on Artificial Intelligence, AAAI 2021
T2 - 35th AAAI Conference on Artificial Intelligence, AAAI 2021
Y2 - 2 February 2021 through 9 February 2021
ER -