Search

We present a novel deep graphical representation that seamlessly merges principles of game theory with laws of statistical mechanics. It performs feature extraction, dimensionality reduction, and pattern classification within a single learning framework. Our approach draws an analogy between neurons in a network and players in a game theory model. Furthermore, each neuron viewed as a classical particle (subject to statistical physics' laws) is mapped to a set of actions representing specific activation value, and neural network layers are conceptualized as games in a sequential cooperative game theory setting. The feed-forward process in deep learning is interpreted as a sequential game, where each game comprises a set of players. During training, neurons are iteratively evaluated and filtered based on their contributions to a payoff function, which is quantified using the Shapley value driven by an energy function. Each set of neurons that significantly contributes to the payoff function forms a strong coalition. These neurons are the only ones permitted to propagate the information forward to the next layers. We applied this methodology to the task of facial age estimation and gender classification. Experimental results demonstrate that our approach outperforms both multi-layer perceptron and convolutional neural network models in terms of efficiency and accuracy.

We present a computational formulation for the approximate version of several variational inequality problems, investigating their computational complexity and establishing PPAD-completeness. Examining applications in computational game theory, we specifically focus on two key concepts: resilient Nash equilibrium, and multi-leader-follower games -- domains traditionally known for the absence of general solutions. In the presence of standard assumptions and relaxation techniques, we formulate problem versions for such games that are expressible in terms of variational inequalities, ultimately leading to proofs of PPAD-completeness.

Conventional federated learning frameworks suffer from several challenges including performance bottlenecks at the central aggregation server, data bias, poor model convergence, and exposure to model poisoning attacks, and limited trust in the centralized infrastructure. In the current paper, a novel game theory-based approach called pFedGame is proposed for decentralized federated learning, best suitable for temporally dynamic networks. The proposed algorithm works without any centralized server for aggregation and incorporates the problem of vanishing gradients and poor convergence over temporally dynamic topology among federated learning participants. The solution comprises two sequential steps in every federated learning round, for every participant. First, it selects suitable peers for collaboration in federated learning. Secondly, it executes a two-player constant sum cooperative game to reach convergence by applying an optimal federated learning aggregation strategy. Experiments performed to assess the performance of pFedGame in comparison to existing methods in decentralized federated learning have shown promising results with accuracy higher than 70% for heterogeneous data.

AI agents will be predictable in certain ways that traditional agents are not. Where and how can we leverage this predictability in order to improve social welfare? We study this question in a game-theoretic setting where one agent can pay a fixed cost to simulate the other in order to learn its mixed strategy. As a negative result, we prove that, in contrast to prior work on pure-strategy simulation, enabling mixed-strategy simulation may no longer lead to improved outcomes for both players in all so-called "generalised trust games". In fact, mixed-strategy simulation does not help in any game where the simulatee's action can depend on that of the simulator. We also show that, in general, deciding whether simulation introduces Pareto-improving Nash equilibria in a given game is NP-hard. As positive results, we establish that mixed-strategy simulation can improve social welfare if the simulator has the option to scale their level of trust, if the players face challenges with both trust and coordination, or if maintaining some level of privacy is essential for enabling cooperation.

Examining the mechanisms underlying the formation and evolution of opinions within real-world social systems, which consist of numerous individuals, can provide valuable insights for effective social functioning and informed business decision making. The focus of our study is on the dynamics of opinions inside a networked multi-agent system. We provide a novel approach called the Game Theory Based Community-Aware Opinion Formation Process (GCAOFP) to accurately represent the co-evolutionary dynamics of communities and opinions in real-world social systems. The GCAOFP algorithm comprises two distinct steps in each iteration. 1) The Community Dynamics Process conceptualizes the process of community formation as a non-cooperative game involving a finite number of agents. Each individual agent aims to maximize their own utility by adopting a response that leads to the most favorable update of the community label. 2) The Opinion Formation Process involves the updating of an individual agent's opinion within a community-aware framework that incorporates bounded confidence. This process takes into account the updated matrix of community members and ensures that an agent's opinion aligns with the opinions of others within their community, within certain defined limits. The present study provides a theoretical proof that under any initial conditions, the aforementioned co-evolutionary dynamics process will ultimately reach an equilibrium state. In this state, both the opinion vector and community member matrix will stabilize after a finite number of iterations. In contrast to conventional opinion dynamics models, the guaranteed convergence of agent opinion within the same community ensures that the convergence of opinions takes place exclusively inside a given community., Comment: 36 pages, 15figures

We consider the automatic online synthesis of black-box test cases from functional requirements specified as automata for reactive implementations. The goal of the tester is to reach some given state, so as to satisfy a coverage criterion, while monitoring the violation of the requirements. We develop an approach based on Monte Carlo Tree Search, which is a classical technique in reinforcement learning for efficiently selecting promising inputs. Seeing the automata requirements as a game between the implementation and the tester, we develop a heuristic by biasing the search towards inputs that are promising in this game. We experimentally show that our heuristic accelerates the convergence of the Monte Carlo Tree Search algorithm, thus improving the performance of testing.

With the continuous advancement of network technology, various emerging complex networking optimization problems opened up a wide range of applications utilizating of game theory. However, since game theory is a mathematical framework, game theory-based solutions often require the experience and knowledge of human experts. Recently, the remarkable advantages exhibited by generative artificial intelligence (GAI) have gained widespread attention. In this article, we propose a novel GAI-enabled game theory solution that combines the powerful reasoning and generation capabilities of GAI to the design and optimization of mobile networking. Specifically, we first outline the game theory and key technologies of GAI, and then explore the advantages of combining GAI with game theory. Then, we briefly review the advantages and limitations of existing research and demonstrate the potential application values of GAI applied to game theory in mobile networking. Subsequently, we develop a game theory framework enabled by large language models (LLMs) to realize this combination, and demonstrate the effectiveness of the proposed framework through a case study in secured UAV networks. Finally, we provide several directions for future extensions.

We consider inverse reinforcement learning problems with concave utilities. Concave Utility Reinforcement Learning (CURL) is a generalisation of the standard RL objective, which employs a concave function of the state occupancy measure, rather than a linear function. CURL has garnered recent attention for its ability to represent instances of many important applications including the standard RL such as imitation learning, pure exploration, constrained MDPs, offline RL, human-regularized RL, and others. Inverse reinforcement learning is a powerful paradigm that focuses on recovering an unknown reward function that can rationalize the observed behaviour of an agent. There has been recent theoretical advances in inverse RL where the problem is formulated as identifying the set of feasible reward functions. However, inverse RL for CURL problems has not been considered previously. In this paper we show that most of the standard IRL results do not apply to CURL in general, since CURL invalidates the classical Bellman equations. This calls for a new theoretical framework for the inverse CURL problem. Using a recent equivalence result between CURL and Mean-field Games, we propose a new definition for the feasible rewards for I-CURL by proving that this problem is equivalent to an inverse game theory problem in a subclass of mean-field games. We outline future directions and applications in human--AI collaboration enabled by our results.

We revive an old lateral-thinking puzzle by Michael Rabin, involving poisons with strange properties. We show that the puzzle admits several unintended solutions that are just as interesting as the intended solution. Analyzing these alternative solutions using game theory yields surprisingly subtle results and several unanswered questions., Comment: 7 pages, to be published in Mathematics Magazine

The application of game theory in cybersecurity enables strategic analysis, adversarial modeling, and optimal decision-making to address security threats' complex and dynamic nature. Previous studies by Abraham et al. and Bi\c{c}er et al. presented various definitions of equilibria to examine the security aspects of games involving multiple parties. Nonetheless, these definitions lack practical and easy-to-use implementations. Our primary contribution is addressing this gap by developing Gamu Blue, an easy-to-use tool with implementations for computing the equilibria definitions including k-resiliency, l-repellence, t-immunity, (l, t)-resistance, and m-stability.

Understanding human behaviour in decision problems and strategic interactions has wide-ranging applications in economics, psychology, and artificial intelligence. Game theory offers a robust foundation for this understanding, based on the idea that individuals aim to maximize a utility function. However, the exact factors influencing strategy choices remain elusive. While traditional models try to explain human behaviour as a function of the outcomes of available actions, recent experimental research reveals that linguistic content significantly impacts decision-making, thus prompting a paradigm shift from outcome-based to language-based utility functions. This shift is more urgent than ever, given the advancement of generative AI, which has the potential to support humans in making critical decisions through language-based interactions. We propose sentiment analysis as a fundamental tool for this shift and take an initial step by analyzing 61 experimental instructions from the dictator game, an economic game capturing the balance between self-interest and the interest of others, which is at the core of many social interactions. Our meta-analysis shows that sentiment analysis can explain human behaviour beyond economic outcomes. We discuss future research directions. We hope this work sets the stage for a novel game theoretical approach that emphasizes the importance of language in human decisions.

Evolutionary Game Theory (EGT) and Artificial Intelligence (AI) are two fields that, at first glance, might seem distinct, but they have notable connections and intersections. The former focuses on the evolution of behaviors (or strategies) in a population, where individuals interact with others and update their strategies based on imitation (or social learning). The more successful a strategy is, the more prevalent it becomes over time. The latter, meanwhile, is centered on machine learning algorithms and (deep) neural networks. It is often from a single-agent perspective but increasingly involves multi-agent environments, in which intelligent agents adjust their strategies based on feedback and experience, somewhat akin to the evolutionary process yet distinct in their self-learning capacities. In light of the key components necessary to address real-world problems, including (i) learning and adaptation, (ii) cooperation and competition, (iii) robustness and stability, and altogether (iv) population dynamics of individual agents whose strategies evolve, the cross-fertilization of ideas between both fields will contribute to the advancement of mathematics of multi-agent learning systems, in particular, to the nascent domain of ``collective cooperative intelligence'' bridging evolutionary dynamics and multi-agent reinforcement learning., Comment: 8 pages, 1 figure

In this paper, we study the highly competitive arena of baby naming. Through making several Extremely Reasonable Assumptions (namely, that parents are myopic, perfectly knowledgeable agents who pick a name based solely on its uniqueness), we create a model which is not only tractable and clean, but also perfectly captures the real world. We then extend our investigation with numerical experiments, as well as analysis of large language model tools. We conclude by discussing avenues for future research., Comment: Accepted at SIGBOVIK 2024

In this paper, we study a theoretical math problem of game theory and calculus of variations in which we minimize a functional involving two players. A general relationship between the optimal strategies for both players is presented, followed by computer analysis as well as polynomial approximation. Nash equilibrium strategies are determined through algebraic manipulation and linear programming. Lastly, a variation of the game is also investigated., Comment: 29 pages, 17 figures

Game theory, as an analytical tool, is frequently utilized to analyze human behavior in social science research. With the high alignment between the behavior of Large Language Models (LLMs) and humans, a promising research direction is to employ LLMs as substitutes for humans in game experiments, enabling social science research. However, despite numerous empirical researches on the combination of LLMs and game theory, the capability boundaries of LLMs in game theory remain unclear. In this research, we endeavor to systematically analyze LLMs in the context of game theory. Specifically, rationality, as the fundamental principle of game theory, serves as the metric for evaluating players' behavior -- building a clear desire, refining belief about uncertainty, and taking optimal actions. Accordingly, we select three classical games (dictator game, Rock-Paper-Scissors, and ring-network game) to analyze to what extent LLMs can achieve rationality in these three aspects. The experimental results indicate that even the current state-of-the-art LLM (GPT-4) exhibits substantial disparities compared to humans in game theory. For instance, LLMs struggle to build desires based on uncommon preferences, fail to refine belief from many simple patterns, and may overlook or modify refined belief when taking actions. Therefore, we consider that introducing LLMs into game experiments in the field of social science should be approached with greater caution., Comment: AAAI 2024

Poker is in the family of imperfect information games unlike other games such as chess, connect four, etc which are perfect information game instead. While many perfect information games have been solved, no non-trivial imperfect information game has been solved to date. This makes poker a great test bed for Artificial Intelligence research. In this paper we firstly compare Game theory optimal poker to Exploitative poker. Secondly, we discuss the intricacies of abstraction techniques, betting models, and specific strategies employed by successful poker bots like Tartanian[1] and Pluribus[6]. Thirdly, we also explore 2-player vs multi-player games and the limitations that come when playing with more players. Finally, this paper discusses the role of machine learning and theoretical approaches in developing winning strategies and suggests future directions for this rapidly evolving field.

The global impact of the COVID-19 pandemic on the livelihoods of people worldwide prompted the implementation of a range of preventive measures at local, national, and international levels. Early in the outbreak, before the vaccine became accessible, voluntary quarantine and social isolation emerged as crucial strategies to curb the spread of infection. In this research, we present a game-theoretic model to elucidate the voluntary disclosure of exposure to infected individuals within communities. By employing a fractional derivative approach to illustrate disease propagation within the compartmental model, we determine the minimum level of voluntary disclosure required to disrupt the chain of transmission and allow the epidemic to fade. Our findings suggest that higher transmission rates and increased perceived severity of infection change the externality of disclosing infected exposure, thereby contributing to a rise in the proportion of individuals opting for quarantine and reducing disease incidence. We estimate behavioral parameters and transmission rates by fitting the model to hospitalized cases in Chile, South America. Results from our paper underscore the potential for public health authorities to influence and regulate voluntary disclosure of infection during emerging outbreaks through effective risk communication, emphasizing the severity of the disease, and providing accurate information about hospital capacity to the public.

An $n$-player game $X$ in normal form can be modeled via undirected discrete graphical models where the discrete random variables represent the players and their state spaces are the set of pure strategies. There exists an edge between the vertices of the graphical model whenever there is a dependency between the associated players. We study the Spohn conditional independence (CI) variety $\mathcal{V}_{X,\mathcal{C}}$, which is the intersection of the independence model $\mathcal{M}_{\mathcal{C}}$ with the Spohn variety of the game $X$. We prove a conjecture by the first author and Sturmfels that $\mathcal{V}_{X,\mathcal{C}}$ is of codimension $n$ in $\mathcal{M}_{\mathcal{C}}$ for a generic game $X$ with binary choices. We show that the set of totally mixed CI equilibria i.e., the restriction of the Spohn CI variety to the open probability simplex is a smooth semialgebraic manifold for a generic game $X$ with binary choices. If the undirected graph is a disjoint union of cliques, we analyze certain algebro-geometric features of Spohn CI varieties and prove affine universality theorems., Comment: 30 pages, 4 figures