A game‐theoretic model for resource allocation with deception and defense efforts

Abstract

This paper develops a strategy for assisting two players in allocating multiple resources in a strategic sequential game. The defender first needs to allocate deception and defense efforts among targets to deceive the attacker and strengthen the target, respectively. Then, the attacker chooses a type of threat and a target to attack. The defender aims at mitigating the possible damage to the targets, whereas the attacker strives to cause maximum damage to the targets. Traditional modeling approaches typically focus only on the defender's homogeneous resource in defense and are not well suited to effectively capture the complex interplay between players. Given scarce resources, a game‐theoretic model is proposed for determining optimal strategies for both players. The key novel features of this model include: (1) the attacker's learning and the defender's counter‐learning efforts are considered; (2) trade‐offs between deception and defense efforts among different targets for the defender are investigated; and (3) sensitive analysis is carried out to see how different parameters can affect the equilibrium results. An illustrative example is presented to demonstrate the procedure of this game‐theoretic model and show its effectiveness. The results can provide additional insights for defense and deception strategies.

Utilizing a maximin optimization approach to maximize system resiliency

“…The literature review demonstrates that, while some work has been done to apply game theory to systems engineering through the design of Systems of Systems with a clear adversary (e.g., a defensive system), little to no work has been done to generalize these concepts to complex systems without an obvious adversary. 34,35 Additionally, while mathematical programing has been applied to system design in several instances, these applications neglect the operating environment as an active (if uncontrollable) element of system design. 20,25,26 The proposed methodology lays out an approach for combining these two methods (mathematical programing and game theory) to ensure the system design is resilient to all operating environments in which the system would be expected to perform.…”

“…Particularly, it is growing in use by systems engineers who "are considering the role of humans in the design and development of complex systems." 34 This has resulted in more insightful System of Systems design and more intuitive and realistic agent based modeling. 35…”

Game-Theoretic and Machine Learning-based Approaches for Defensive Deception: A Survey

“…Hypergame uses a subgame to model each player's different view on the game and corresponding strategies under the subgame. In addition, Zhang et al [153] and Xu and Zhuang [142] used Subgame Perfect Nash Equilibrium (SPNE) [94] (i.e., a Nash equilibrium in every subgame) to deal with the resource allocation for both the defender and attacker in non-hypergames.…”

“…• Utility [14,15,13,12,37,49,65,73,75,81,96,97,99,101,100,106,107,111,114,122,132,133,137,141,153,155]:…”

“…As a result, some of these studies considered the impact to normal as another cost of deception technology [114,137]. A defender's expected utility is a common metric to be maximized as the system's objective, as shown in game-theoretic or ML-based DD techniques in [5,7,65,73,75,81,85,96,100,106,107,111,122,132,133,141,153,155].…”