Machine learning processes, e.g. ''learning in games'', can be viewed as non-linear dynamical systems. In general, such systems exhibit a wide spectrum of behaviors, ranging from stability/recurrence to the undesirable phenomena of chaos (or ''butterfly effect''). Chaos captures sensitivity of round-off errors and can severely affect predictability and reproducibility of ML systems, but AI/ML community's understanding of it remains rudimentary. It has a lot out there that await exploration. Recently, Cheung and Piliouras employed volume-expansion argument to show that Lyapunov chaos occurs in the cumulative payoff space, when some popular learning algorithms, including Multiplicative Weights Update (MWU), Follow-the-Regularized-Leader (FTRL) and Optimistic MWU (OMWU), are used in several subspaces of games, e.g. zero-sum, coordination or graphical constant-sum games. It is natural to ask: can these results generalize to much broader families of games? We take on a game decomposition approach and answer the question affirmatively. Among other results, we propose a notion of ''matrix domination'' and design a linear program, and use them to characterize bimatrix games where MWU is Lyapunov chaotic almost everywhere. Such family of games has positive Lebesgue measure in the bimatrix game space, indicating that chaos is a substantial issue of learning in games. For multi-player games, we present a local equivalence of volume change between general games and graphical games, which is used to perform volume and chaos analyses of MWU and OMWU in potential games.
We study a novel control problem in the context of network coordination games: the individuation of the smallest set of players capable of driving the system, globally, from one Nash equilibrium to another one. Our main contribution is the design of a randomized algorithm based on a time-reversible Markov chain with provable convergence garantees.
This paper aims to design a distributed coordination algorithm for solving a multi-agent decision problem with a hierarchical structure. The primary goal is to search the Nash equilibrium of a noncooperative game such that each player has no incentive to deviate from the equilibrium under its private objective. Meanwhile, the agents can coordinate to optimize the social cost within the set of Nash equilibria of the underlying game. Such an optimal Nash equilibrium problem can be modeled as a distributed optimization problem with variational inequality constraints. We consider the scenario wher…
This paper deals with the complexity of the problem of computing a pure Nash equilibrium for discrete preference games and network coordination games beyond $O(\log n)$-treewidth and tree metric spaces. First, we estimate the number of iterations of the best response dynamics for a discrete preference game on a discrete metric space with at least three strategies. Second, we present a sufficient condition that we have a polynomial-time algorithm to find a pure Nash equilibrium for a discrete preference game on a grid graph. Finally, we discuss the complexity of finding a pure Nash equilibrium …
In this study, we formulate positive and negative externalities caused by changes in the supply of shared vehicles as ride sharing games. The study aims to understand the price of anarchy (PoA) and its improvement via a coordination technique in ride sharing games. A critical question is whether ride sharing games exhibit a pure Nash equilibrium (pNE) since the PoA bound assumes it. Our result shows a sufficient condition for a ride sharing game to have a finite improvement property and a pNE similar to potential games. This is the first step to analyze PoA bound and its improvement by coordin…
We study natural strategic games on directed graphs, which capture the idea of coordination in the absence of globally common strategies. We show that these games do not need to have a pure Nash equilibrium and that the problem of determining their existence is NP-complete. The same holds for strong equilibria. We also exhibit some classes of games for which strong equilibria exist and prove that a strong equilibrium can then be found in linear time.
This paper investigates repeated win-lose coordination games (WLC-games). We analyse which protocols are optimal for these games covering both the worst case and average case scenarios, i,e., optimizing the guaranteed and expected coordination times. We begin by analysing Choice Matching Games (CM-games) which are a simple yet fundamental type of WLC-games, where the goal of the players is to pick the same choice from a finite set of initially indistinguishable choices. We give a complete classification of optimal expected and guaranteed coordination times in two-player CM-games and show that …
Whilst network coordination games and network anti-coordination games have received a considerable amount of attention in the literature, network games with coexisting coordinating and anti-coordinating players are known to exhibit more complex behaviors. In fact, depending on the network structure, such games may even fail to have pure-strategy Nash equilibria. An example is represented by the well-known matching pennies (discoordination) game. In this work, we first provide graph-theoretic conditions for the existence of pure-strategy Nash equilibria in mixed network coordination/anti-coordi…
Sharing unused vehicles is one practical solution for traffic congestion. We propose an advanced vehicle-sharing service that maximizes the sharing of vehicles and improves traffic efficiency by coordinating user trips via an information system. We formulate ride-sharing games that model externalities in vehicle sharing caused by insufficient vehicle supply. We show how Bayes correlated equilibrium can coordinate players in ride-sharing games and verify the resultant improvement in the price of anarchy.
Logit Dynamics [Blume, Games and Economic Behavior, 1993] are randomized best response dynamics for strategic games: at every time step a player is selected uniformly at random and she chooses a new strategy according to a probability distribution biased toward strategies promising higher payoffs. This process defines an ergodic Markov chain, over the set of strategy profiles of the game, whose unique stationary distribution is the long-term equilibrium concept for the game. However, when the mixing time of the chain is large (e.g., exponential in the number of players), the stationary distrib…
This paper investigates repeated win-lose coordination games (WLC-games). We analyse which protocols are optimal for these games, covering both the worst case and average case scenarios, i,e., optimizing the guaranteed and expected coordination times. We begin by analysing Choice Matching Games (CM-games) which are a simple yet fundamental type of WLC-games, where the goal of the players is to pick the same choice from a finite set of initially indistinguishable choices. We give a fully complete classification of optimal expected and guaranteed coordination times in two-player CM-games and sho…
We consider the control of decentralized learning dynamics for agents in an anti-coordination network game. In the anti-coordination network game, there is a preferred action in the absence of neighbors' actions, and the utility an agent receives from the preferred action decreases as more of its neighbors select the preferred action, potentially causing the agent to select a less desirable action. The decentralized dynamics that is based on the iterated elimination of dominated strategies converge for the considered game. Given a convergent action profile, we measure anti-coordination by the …
Motivated by understanding non-strict and strict pure strategy equilibria in network anti-coordination games, we define notions of stable and, respectively, strictly stable colorings in graphs. We characterize the cases when such colorings exist and when the decision problem is NP-hard. These correspond to finding pure strategy equilibria in the anti-coordination games, whose price of anarchy we also analyze. We further consider the directed case, a generalization that captures both coordination and anti-coordination. We prove the decision problem for non-strict equilibria in directed graphs i…
We introduce and motivate the study of hypergraphical clustering games of mis-coordination. For two specific variants we prove the existence of a pure Nash equilibrium and provide bounds on the price of anarchy as a function of the cardinality of the action set and the size of the hyperedges.
We study the role of local effects and finite size effects in reaching coordination and in equilibrium selection in different types of two-player coordination games. We investigate three update rules -- the replicator dynamics (RD), the best response (BR), and the unconditional imitation (UI) -- for coordination games on random graphs. Local effects turn out to me significantly more important for the UI update rule. For the pure coordination game with two equivalent strategies we find a transition from a disordered state to a state of full coordination for a critical value of the network conne…
Lund University（隆德大学）；The University of Texas at Austin（得克萨斯大学奥斯汀分校）；Georgia Institute of Technology（佐治亚理工学院）
We consider the problem of distributed convergence to efficient outcomes in coordination games through dynamics based on aspiration learning. Under aspiration learning, a player continues to play an action as long as the rewards received exceed a specified aspiration level. Here, the aspiration level is a fading memory average of past rewards, and these levels also are subject to occasional random perturbations. A player becomes dissatisfied whenever a received reward is less than the aspiration level, in which case the player experiments with a probability proportional to the degree of dissat…
We study the problem of stochastic stability for evolutionary dynamics under the logit choice rule. We consider general classes of coordination games, symmetric or asymmetric, with an arbitrary number of strategies, which satisfies the marginal bandwagon property (i.e., there is positive feedback to coordinate). Our main result is that the most likely evolutionary escape paths from a status quo convention consist of a series of identical mistakes. As an application of our result, we show that the Nash bargaining solution arises as the long run convention for the evolutionary Nash demand game u…
Communities typically capture homophily as people of the same community share many common features. This paper is motivated by the problem of community detection in social networks, as it can help improve our understanding of the network topology and the spread of information. Given the selfish nature of humans to align with like-minded people, we employ game theoretic models and algorithms to detect communities in this paper. Specifically, we employ coordination games to represent interactions between individuals in a social network. We represent the problem of community detection as a graph …
We present new coordination mechanisms for scheduling selfish jobs on $m$ unrelated machines. A coordination mechanism aims to mitigate the impact of selfishness of jobs on the efficiency of schedules by defining a local scheduling policy on each machine. The scheduling policies induce a game among the jobs and each job prefers to be scheduled on a machine so that its completion time is minimum given the assignments of the other jobs. We consider the maximum completion time among all jobs as the measure of the efficiency of schedules. The approximation ratio of a coordination mechanism quantif…
We consider polymatrix coordination games with individual preferences where every player corresponds to a node in a graph who plays with each neighbor a separate bimatrix game with non-negative symmetric payoffs. In this paper, we study $α$-approximate $k$-equilibria of these games, i.e., outcomes where no group of at most $k$ players can deviate such that each member increases his payoff by at least a factor $α$. We prove that for $α\ge 2$ these games have the finite coalitional improvement property (and thus $α$-approximate $k$-equilibria exist), while for $α< 2$ this property does not hold.…
Stockholm University（斯德哥尔摩大学）；University of Helsinki（赫尔辛基大学）；Tampere University（坦佩雷大学）；ENS Paris-Saclay（巴黎萨克雷高等师范学校）
We study pure coordination games where in every outcome, all players have identical payoffs, 'win' or 'lose'. We identify and discuss a range of 'purely rational principles' guiding the reasoning of rational players in such games and analyze which classes of coordination games can be solved by such players with no preplay communication or conventions. We observe that it is highly nontrivial to delineate a boundary between purely rational principles and other decision methods, such as conventions, for solving such coordination games.
Coordination mechanisms aim to mitigate the impact of selfishness when scheduling jobs to different machines. Such a mechanism defines a scheduling policy within each machine and naturally induces a game among the selfish job owners. The desirable properties of a coordination mechanism includes simplicity in its definition and efficiency of the outcomes of the induced game. We present a broad class of coordination mechanisms for unrelated machine scheduling that are simple to define and we identify one of its members (mechanism \Dcoord) that is superior to all known mechanisms. \Dcoord\ induce…
How does information regarding an adversary's intentions affect optimal system design? This paper addresses this question in the context of graphical coordination games where an adversary can indirectly influence the behavior of agents by modifying their payoffs. We study a situation in which a system operator must select a graph topology in anticipation of the action of an unknown adversary. The designer can limit her worst-case losses by playing a security strategy, effectively planning for an adversary which intends maximum harm. However, fine-grained information regarding the adversary's i…
We study strategic games on weighted directed graphs, in which the payoff of a player is defined as the sum of the weights on the edges from players who chose the same strategy, augmented by a fixed non-negative integer bonus for picking a given strategy. These games capture the idea of coordination in the absence of globally common strategies. We identify natural classes of graphs for which finite improvement or coalition-improvement paths of polynomial length always exist, and, as a consequence, a (pure) Nash equlibrium or a strong equilibrium can be found in polynomial time. The considered …
We study the population genetics of Evolution in the important special case of weak selection, in which all fitness values are assumed to be close to one another. We show that in this regime natural selection is tantamount to the multiplicative updates game dynamics in a coordination game between genes. Importantly, the utility maximized in this game, as well as the amount by which each allele is boosted, is precisely the allele's mixability, or average fitness, a quantity recently proposed in [1] as a novel concept that is crucial in understanding natural selection under sex, thus providing a…
The ad hoc coordination problem is to design an autonomous agent which is able to achieve optimal flexibility and efficiency in a multiagent system with no mechanisms for prior coordination. We conceptualise this problem formally using a game-theoretic model, called the stochastic Bayesian game, in which the behaviour of a player is determined by its private information, or type. Based on this model, we derive a solution, called Harsanyi-Bellman Ad Hoc Coordination (HBA), which utilises the concept of Bayesian Nash equilibrium in a planning procedure to find optimal actions in the sense of Bel…
We introduce natural strategic games on graphs, which capture the idea of coordination in a local setting. We study the existence of equilibria that are resilient to coalitional deviations of unbounded and bounded size (i.e., strong equilibria and k-equilibria respectively). We show that pure Nash equilibria and 2-equilibria exist, and give an example in which no 3-equilibrium exists. Moreover, we prove that strong equilibria exist for various special cases. We also study the price of anarchy (PoA) and price of stability (PoS) for these solution concepts. We show that the PoS for strong equili…
Strategic Delay and Coordination Efficiency in Global Games
King Abdullah University of Science and Technology（阿卜杜拉国王科技大学）；University of Illinois Urbana Champaign（伊利诺伊大学厄巴纳-香槟分校）；Florida State University（佛罗里达州立大学）
We consider the charge scheduling coordination of a fleet of plug-in electric vehicles, developing a hybrid decision-making framework for efficient and profitable usage of the distribution grid. Each charging dynamics, affected by the aggregate behavior of the whole fleet, is modelled as an inter-dependent, mixed-logical-dynamical system. The coordination problem is formalized as a generalized mixed-integer aggregative potential game, and solved via semi-decentralized implementation of a sequential best-response algorithm that leads to an approximated equilibrium of the game.
This paper introduces a differentially private (DP) mechanism to protect the information exchanged during the coordination of sequential and interdependent markets. This coordination represents a classic Stackelberg game and relies on the exchange of sensitive information between the system agents. The paper is motivated by the observation that the perturbation introduced by traditional DP mechanisms fundamentally changes the underlying optimization problem and even leads to unsatisfiable instances. To remedy such limitation, the paper introduces the Privacy-Preserving Stackelberg Mechanism (P…
We study strategic games on weighted directed graphs, where the payoff of a player is defined as the sum of the weights on the edges from players who chose the same strategy augmented by a fixed non-negative bonus for picking a given strategy. These games capture the idea of coordination in the absence of globally common strategies. Prior work shows that the problem of determining the existence of a pure Nash equilibrium for these games is NP-complete already for graphs with all weights equal to one and no bonuses. However, for several classes of graphs (e.g. DAGs and cliques) pure Nash equili…
To achieve an optimal outcome in many situations, agents need to choose distinct actions from one another. This is the case notably in many resource allocation problems, where a single resource can only be used by one agent at a time. How shall a designer of a multi-agent system program its identical agents to behave each in a different way? From a game theoretic perspective, such situations lead to undesirable Nash equilibria. For example consider a resource allocation game in that two players compete for an exclusive access to a single resource. It has three Nash equilibria. The two pure-str…
In this work we assess the role played by the dynamical adaptation of the interactions network, among agents playing Coordination Games, in reaching global coordination and in the equilibrium selection. Specifically, we analyze a coevolution model that couples the changes in agents' actions with the network dynamics, so that while agents play the game, they are able to sever some of their current connections and connect with others. We focus on two update rules: Replicator Dynamics (RD) and Unconditional Imitation (UI). We investigate a Pure Coordination Game (PCG), in which choices are equiva…
Coordination and cooperation between humans and autonomous agents in cooperative games raises interesting questions of human decision making and behaviour changes. Here we report our findings from a group formation game in a small-world network of different mixes of human and agent players, aiming to achieve connected clusters of the same colour by swapping places with neighbouring players using non-overlapping information. In the experiments the human players are incentivized by rewarding to prioritize their own cluster while the model of agents' decision making is derived from our previous e…
In this work, we develop a reinforcement learning protocol for a multiagent coordination task in a discrete state and action space: an iterated prisoner's dilemma game extended into a team based, winner-take all tournament, which forces the agents to collude in order to maximize their reward. By disallowing extra communication channels, the agents are forced to embed their coordination strategy into their actions in the prisoner's dilemma game. We develop a representation of the iterated prisoners dilemma that makes it amenable to Q-learning. We find that the reinforcement learning strategy is…
Forgetting Our Way to Shared Meaning: Effects of Forgetting on Conceptual Alignment in a Non-Partnership Coordination Game
Institute for Complex Social Dynamics, Carnegie Mellon University（复杂社会动态研究所，卡内基梅隆大学）；Department of Computer Science, Carleton University（计算机科学系，卡尔顿大学）；Department of Cognitive Science, Carleton University（认知科学系，卡尔顿大学）
The stag hunt (or assurance game) is a simple game that has been used as a prototype of a variety of social coordination problems (ranging from the social contract to the adoption of technical standards). Players have the option to either use a superior cooperative strategy whose payoff depends on the other players' choices or use an inferior strategy whose payoff is independent of what other players do; the cooperative strategy may incur a loss if sufficiently many other players do not cooperate. Stag hunts have two (strict) pure Nash equilibria, namely, universal cooperation and universal de…
We consider games of strategic substitutes and strategic complements on networks. We introduce two different evolutionary dynamics in order to refine their multiplicity of equilibria, and we analyse the system through a mean field approach. We find that for the best-shot game, taken as a model for substitutes, a replicator-like dynamics does not lead to Nash equilibria, whereas it leads to unique equilibria (full cooperation or full defection, depending on the initial condition and the game parameter) for complements, represented by a coordination game. On the other hand, when the dynamics bec…
Cracow University of Economics（克拉科夫经济大学）；Chulalongkorn University（朱拉隆功大学）；Indiana University-Purdue University Indianapolis（印第安纳大学-普渡大学印第安纳波利斯分校）；Singapore University of Technology and Design（新加坡科技设计大学）
We study the emergence of chaotic behavior of Follow-the-Regularized Leader (FoReL) dynamics in games. We focus on the effects of increasing the population size or the scale of costs in congestion games, and generalize recent results on unstable, chaotic behaviors in the Multiplicative Weights Update dynamics to a much larger class of FoReL dynamics. We establish that, even in simple linear non-atomic congestion games with two parallel links and any fixed learning rate, unless the game is fully symmetric, increasing the population size or the scale of costs causes learning dynamics to become u…
