The present survey aims at presenting the current machine learning techniques employed in security games domains. Specifically, we focused on papers and works developed by the Teamcore of University of Southern California, which deepened different directions in this field. After a brief introduction on Stackelberg Security Games (SSGs) and the poaching setting, the rest of the work presents how to model a boundedly rational attacker taking into account her human behavior, then describes how to face the problem of having attacker's payoffs not defined and how to estimate them and, finally, presents how online learning techniques have been exploited to learn a model of the attacker.
University of California, Berkeley（加利福尼亚大学伯克利分校）；Purdue University（普渡大学）；The University of Texas at Austin（得克萨斯大学奥斯汀分校）
In a Stackelberg game, a leader commits to a randomized strategy, and a follower chooses their best strategy in response. We consider an extension of a standard Stackelberg game, called a discrete-time dynamic Stackelberg game, that has an underlying state space that affects the leader's rewards and available strategies and evolves in a Markovian manner depending on both the leader and follower's selected strategies. Although standard Stackelberg games have been utilized to improve scheduling in security domains, their deployment is often limited by requiring complete information of the follow…
We study the computational complexity of finding Stackelberg Equilibria in general-sum games, where the set of pure strategies of the leader and the followers are exponentially large in a natrual representation of the problem. In \emph{zero-sum} games, the notion of a Stackelberg equilibrium coincides with the notion of a \emph{Nash Equilibrium}~\cite{korzhyk2011stackelberg}. Finding these equilibrium concepts in zero-sum games can be efficiently done when the players have polynomially many pure strategies or when (in additional to some structural properties) a best-response oracle is availabl…
The behavior of no-regret learning algorithms is well understood in two-player min-max (i.e, zero-sum) games. In this paper, we investigate the behavior of no-regret learning in min-max games with dependent strategy sets, where the strategy of the first player constrains the behavior of the second. Such games are best understood as sequential, i.e., min-max Stackelberg, games. We consider two settings, one in which only the first player chooses their actions using a no-regret algorithm while the second player best responds, and one in which both players use no-regret algorithms. For the former…
Stackelberg equilibrium is a solution concept that describes optimal strategies to commit: Player 1 (the leader) first commits to a strategy that is publicly announced, then Player 2 (the follower) plays a best response to the leader's commitment. We study the problem of computing Stackelberg equilibria in sequential games with finite and indefinite horizons, when players can play history-dependent strategies. Using the alternate formulation called strategies with memory, we establish that strategy profiles with polynomial memory size can be described efficiently. We prove that there exist a p…
Robust Reinforcement Learning (RL) focuses on improving performances under model errors or adversarial attacks, which facilitates the real-life deployment of RL agents. Robust Adversarial Reinforcement Learning (RARL) is one of the most popular frameworks for robust RL. However, most of the existing literature models RARL as a zero-sum simultaneous game with Nash equilibrium as the solution concept, which could overlook the sequential nature of RL deployments, produce overly conservative agents, and induce training instability. In this paper, we introduce a novel hierarchical formulation of ro…
Real world applications such as economics and policy making often involve solving multi-agent games with two unique features: (1) The agents are inherently asymmetric and partitioned into leaders and followers; (2) The agents have different reward functions, thus the game is general-sum. The majority of existing results in this field focuses on either symmetric solution concepts (e.g. Nash equilibrium) or zero-sum games. It remains open how to learn the Stackelberg equilibrium -- an asymmetric analog of the Nash equilibrium -- in general-sum games efficiently from noisy samples. This paper ini…
In this paper, we employ a hypergame framework to analyze the single-leader-multiple-followers (SLMF) Stackelberg security game with two typical misinformed situations: misperception and deception. We provide a stability criterion with the help of hyper Nash equilibrium (HNE) to investigate both strategic stability and cognitive stability of equilibria in SLMF games with misinformation. In fact, we find mild stable conditions such that the equilibria with misperception and deception can become HNE. Moreover, we discuss the robustness of the equilibria to reveal whether players have the ability…
Computation of Stackelberg Equilibria of Finite Sequential Games
Czech Technical University in Prague（布拉格捷克理工大学）；Hebrew University of Jerusalem（耶路撒冷希伯来大学）；Aarhus University（奥胡斯大学）；IT-University of Copenhagen（哥本哈根信息技术大学）
Equilibrium refinements are important in extensive-form (i.e., tree-form) games, where they amend weaknesses of the Nash equilibrium concept by requiring sequential rationality and other beneficial properties. One of the most attractive refinement concepts is quasi-perfect equilibrium. While quasi-perfection has been studied in extensive-form games, it is poorly understood in Stackelberg settings---that is, settings where a leader can commit to a strategy---which are important for modeling, for example, security games. In this paper, we introduce the axiomatic definition of quasi-perfect Stack…
Zero-sum stochastic games have found important applications in a variety of fields, from machine learning to economics. Work on this model has primarily focused on the computation of Nash equilibrium due to its effectiveness in solving adversarial board and video games. Unfortunately, a Nash equilibrium is not guaranteed to exist in zero-sum stochastic games when the payoffs at each state are not convex-concave in the players' actions. A Stackelberg equilibrium, however, is guaranteed to exist. Consequently, in this paper, we study zero-sum stochastic Stackelberg games. Going beyond known exis…
This article investigates the anti-jamming communications problem in wireless networks from a Stackelberg game perspective. By exploring and analyzing the inherent characteristics of the anti-jamming problem, we present and discuss some technical challenges and fundamental requirements to address them. To be specific, the adversarial characteristic, incomplete information constraints, dynamics, uncertainty, dense deployment, and heterogeneous feature bring technical challenges to anti-jamming communications in wireless networks. Then, for the purpose of improving system performance, four requi…
In this paper, we study the notion of adversarial Stackelberg value for two-player non-zero sum games played on bi-weighted graphs with the mean-payoff and the discounted sum functions. The adversarial Stackelberg value of Player 0 is the largest value that Player 0 can obtain when announcing her strategy to Player 1 which in turn responds with any of his best response. For the mean-payoff function, we show that the adversarial Stackelberg value is not always achievable but epsilon-optimal strategies exist. We show how to compute this value and prove that the associated threshold problem is in…
Optimizing strategic decisions (a.k.a. computing equilibrium) is key to the success of many non-cooperative multi-agent applications. However, in many real-world situations, we may face the exact opposite of this game-theoretic problem -- instead of prescribing equilibrium of a given game, we may directly observe the agents' equilibrium behaviors but want to infer the underlying parameters of an unknown game. This research question, also known as inverse game theory, has been studied in multiple recent works in the context of Stackelberg games. Unfortunately, existing works exhibit quite negat…
Computational advertising has been studied to design efficient marketing strategies that maximize the number of acquired customers. In an increased competitive market, however, a market leader (a leader) requires the acquisition of new customers as well as the retention of her loyal customers because there often exists a competitor (a follower) who tries to attract customers away from the market leader. In this paper, we formalize a new model called the Stackelberg budget allocation game with a bipartite influence model by extending a budget allocation problem over a bipartite graph to a Stack…
Min-max optimization problems (i.e., min-max games) have attracted a great deal of attention recently as their applicability to a wide range of machine learning problems has become evident. In this paper, we study min-max games with dependent strategy sets, where the strategy of the first player constrains the behavior of the second. Such games are best understood as sequential, i.e., Stackelberg, games, for which the relevant solution concept is Stackelberg equilibrium, a generalization of Nash. One of the most popular algorithms for solving min-max games is gradient descent ascent (GDA). We …
As an effective approach to thwarting advanced attacks, moving target defense (MTD) has been applied to various domains. Previous works on MTD, however, mainly focus on deciding the sequence of system configurations to be used and have largely ignored the equally important timing problem. Given that both the migration cost and attack time vary over system configurations, it is crucial to jointly optimize the spatial and temporal decisions in MTD to better protect the system from persistent threats. In this work, we propose a Stackelberg game model for MTD where the defender commits to a joint …
Academy of Mathematics and Systems Science, Chinese Academy of Sciences（中国科学院数学与系统科学研究院）；University of Chinese Academy of Sciences（中国科学院大学）
Adversarial deep learning is to train robust DNNs against adversarial attacks, which is one of the major research focuses of deep learning. Game theory has been used to answer some of the basic questions about adversarial deep learning such as the existence of a classifier with optimal robustness and the existence of optimal adversarial samples for a given class of classifiers. In most previous work, adversarial deep learning was formulated as a simultaneous game and the strategy spaces are assumed to be certain probability distributions in order for the Nash equilibrium to exist. But, this as…
University of Maryland（马里兰大学）；Montgomery Blair High School（蒙哥马利布莱尔高中）；Wootton High School（伍顿高中）
Reinforcement Learning (RL) algorithms have been successfully applied to real world situations like illegal smuggling, poaching, deforestation, climate change, airport security, etc. These scenarios can be framed as Stackelberg security games (SSGs) where defenders and attackers compete to control target resources. The algorithm's competency is assessed by which agent is controlling the targets. This review investigates modeling of SSGs in RL with a focus on possible improvements of target representations in RL algorithms.
Stackelberg Games are gaining importance in the last years due to the raise of Adversarial Machine Learning (AML). Within this context, a new paradigm must be faced: in classical game theory, intervening agents were humans whose decisions are generally discrete and low dimensional. In AML, decisions are made by algorithms and are usually continuous and high dimensional, e.g. choosing the weights of a neural network. As closed form solutions for Stackelberg games generally do not exist, it is mandatory to have efficient algorithms to search for numerical solutions. We study two different proced…
Stackelberg equilibrium is a solution concept in two-player games where the leader has commitment rights over the follower. In recent years, it has become a cornerstone of many security applications, including airport patrolling and wildlife poaching prevention. Even though many of these settings are sequential in nature, existing techniques pre-compute the entire solution ahead of time. In this paper, we present a theoretically sound and empirically effective way to apply search, which leverages extra online computation to improve a solution, to the computation of Stackelberg equilibria in ge…
Stackelberg equilibria have become increasingly important as a solution concept in computational game theory, largely inspired by practical problems such as security settings. In practice, however, there is typically uncertainty regarding the model about the opponent. This paper is, to our knowledge, the first to investigate Stackelberg equilibria under uncertainty in extensive-form games, one of the broadest classes of game. We introduce robust Stackelberg equilibria, where the uncertainty is about the opponent's payoffs, as well as ones where the opponent has limited lookahead and the uncert…
In this paper, we consider a discrete-time stochastic Stackelberg game with a single leader and multiple followers. Both the followers and the leader together have conditionally independent private types, conditioned on action and previous state, that evolve as controlled Markov processes. The objective is to compute the stochastic Stackelberg equilibrium of the game where the leader commits to a dynamic strategy. Each follower's strategy is the best response to the leader's strategies and other followers' strategies while the each leader's strategy is optimum given the followers play the best…
We propose a two-layer, semi-decentralized algorithm to compute a local solution to the Stackelberg equilibrium problem in aggregative games with coupling constraints. Specifically, we focus on a single-leader, multiple-follower problem, and after equivalently recasting the Stackelberg game as a mathematical program with complementarity constraints (MPCC), we iteratively convexify a regularized version of the MPCC as inner problem, whose solution generates a sequence of feasible descent directions for the original MPCC. Thus, by pursuing a descent direction at every outer iteration, we establi…
In this paper, we consider a discrete-time Stackelberg mean field game with a finite number of leaders, a finite number of major followers and an infinite number of minor followers. The leaders and the followers each observe types privately that evolve as conditionally independent controlled Markov processes. The leaders are of "Stackelberg" kind which means they commit to a dynamic policy. We consider two types of followers: major and minor, each with a private type. All the followers best respond to the policies of the Stackelberg leaders and each other. Knowing that the followers would play…
Multi-robot cooperation requires agents to make decisions that are consistent with the shared goal without disregarding action-specific preferences that might arise from asymmetry in capabilities and individual objectives. To accomplish this goal, we propose a method named SLiCC: Stackelberg Learning in Cooperative Control. SLiCC models the problem as a partially observable stochastic game composed of Stackelberg bimatrix games, and uses deep reinforcement learning to obtain the payoff matrices associated with these games. Appropriate cooperative actions are then selected with the derived Stac…
Strong Stackelberg equilibrium (SSE) is the standard solution concept of Stackelberg security games. As opposed to the weak Stackelberg equilibrium (WSE), the SSE assumes that the follower breaks ties in favor of the leader and this is widely acknowledged and justified by the assertion that the defender can often induce the attacker to choose a preferred action by making an infinitesimal adjustment to her strategy. Unfortunately, in security games with resource assignment constraints, the assertion might not be valid; it is possible that the defender cannot induce the desired outcome. As a res…
The field of cybersecurity has mostly been a cat-and-mouse game with the discovery of new attacks leading the way. To take away an attacker's advantage of reconnaissance, researchers have proposed proactive defense methods such as Moving Target Defense (MTD). To find good movement strategies, researchers have modeled MTD as leader-follower games between the defender and a cyber-adversary. We argue that existing models are inadequate in sequential settings when there is incomplete information about a rational adversary and yield sub-optimal movement strategies. Further, while there exists an ar…
In a stochastic Stackelberg asymmetric security game, the strong Stackelberg equilibrium (SSE) strategy is a popular option for the defender to get the highest utility against an attacker with the best response (BR) strategy. However, the attacker may be a boundedly rational player, who adopts a combination of the BR strategy and a fixed stubborn one. In such a condition, the SSE strategy may not maintain the defensive performance due to the stubborn element. In this paper, we focus on how the defender can adopt the unilateral-control zero-determinate (ZD) strategy to confront the boundedly ra…
