arXiv subjects

Muhammed O. Sayin

Publications and source records attributed to Muhammed O. Sayin.

At least 19 recordsLinked to original sources

arXiv2026

Decentralized Optimal Equilibrium Learning Over Dynamic Networks

This paper studies decentralized learning of socially optimal equilibria in finite normal-form games over dynamic communication networks. Each agent observes only its own realized payoffs, does not know the game a priori, and can communicate only with time-varying neighbors using low-bandwidth messages. We propose networked decentralized optimal equilibrium learning dynamics in which agents generate randomized semantic content/discontent signals from local payoff comparisons and exchange time-stamped time-stacked tables rather than raw actions, payoff information or local estimates/parameters. The method combines table fusion with temporal majority reconstruction to mitigate dynamic communication while preserving fully decentralized operation. We establish finite-time logarithmic regret guarantees, with an in-phase exploration perturbation, for optimal equilibrium selection under utilitarian and proportional-fair social welfare objectives. Simulation results further show that the proposed approach can effectively select socially desirable equilibria over dynamic communication networks.

cs.GT↗

arXiv2026

A Distributionally Robust Multi-agent Reinforcement Learning Framework for Intelligent Intersection Control

Multi-agent reinforcement learning (MARL) has emerged as a promising approach for traffic signal control. However, standard MARL policies typically optimize for expected returns under nominal conditions, leaving them highly vulnerable to spatial-temporal demand shifts and catastrophic congestion under adverse scenarios. To address this critical limitation, this paper proposes an algorithm-agnostic Distributionally Robust (DR) MARL framework integrating an adaptive Contextual-Bandit Worst-Case Estimator (CB-WCE). Operating on a slower timescale, the CB-WCE co-evolves with the traffic controllers by dynamically generating adversarial demand mixtures during training. This steers the learning process to fortify policies against bottleneck scenarios without requiring modifications to the underlying MARL architectures. The framework is evaluated across value-based, actor-critic, and policy-gradient methods on both a synthetic 5x5 grid and a heterogeneous Monaco City network. Empirical results demonstrate that the DR framework prevents unbounded queue growth and profoundly enhances both worst-case robustness and average-case efficiency. Notably, for the Proximal Policy Optimization (PPO) architecture in the Monaco environment, on average, robust retraining reduced the worst-case queue length by 74.39% and improved the average-case network-wide queue length by 75.45%. Furthermore, the retrained policies exhibit strong zero-shot generalization to unseen traffic distributions, highlighting the framework's scalability and potential for resilient real-world urban deployment.

eess.SY↗

arXiv2026

Controlling Traffic without Tolls: A Non-Monetary Framework for Autonomous Intersections

The increasing complexity of urban transportation systems, driven by connected and automated vehicles, calls for new modeling paradigms and scalable control strategies. We propose a non-monetary control framework that leverages autonomous intersection management to influence routing decisions without tolls. The approach uses timestamp-based scheduling adjustments at roadside units (RSUs) to introduce path-dependent delays or advancements, steering traffic toward socially efficient flows. We develop a hierarchical architecture that separates real-time intersection control from network-level coordination. The resulting model admits a congestion-game formulation with path-dependent node costs. We establish the existence and essential uniqueness of equilibrium flows, eliminating ambiguities due to multiple equilibria and enabling a scalable and tractable bilevel optimization formulation for system-level incentive design. Experiments on the Sioux Falls network show that the proposed approach reduces the efficiency gap between user equilibrium and system-optimal flows by up to 71% under realistic constraints. These results demonstrate the potential of non-monetary, infrastructure-light control for next-generation intelligent transportation and urban mobility systems.

cs.GT↗

arXiv2026

Decentralized Optimal Equilibrium Learning in Stochastic Games via Single-bit Feedback

We study decentralized equilibrium selection in stochastic games under severe information and communication constraints. In such settings, convergence to equilibrium alone is insufficient, as stochastic games typically admit many equilibria with markedly different welfare properties. We address decentralized optimal equilibrium selection, where agents coordinate on equilibria that optimize a designer-specified social welfare objective while allowing heterogeneous tolerance to deviations from strict best responses. Agents observe only the global state trajectory and their realized rewards, and exchange a single randomized bit of feedback per agent per round. This semantic content/discontent signaling mechanism implicitly aligns decentralized learning dynamics with the global welfare objective. We develop explore-and-commit and online variants applicable to general stochastic games, accommodating heterogeneous model-based or model-free methods for solving the induced Markov decision processes, and establish explicit finite-time regret guarantees, showing logarithmic expected regret under mild conditions.

cs.GT↗

arXiv2026

Actor-Dual-Critic Dynamics for Zero-sum and Identical-Interest Stochastic Games

We propose a novel independent and payoff-based learning framework for stochastic games that is model-free, game-agnostic, and gradient-free. The learning dynamics follow a best-response-type actor-critic architecture, where agents update their strategies (actors) using feedback from two distinct critics: a fast critic that intuitively responds to observed payoffs under limited information, and a slow critic that deliberatively approximates the solution to the underlying dynamic programming problem. Crucially, the learning process relies on non-equilibrium adaptation through smoothed best responses to observed payoffs. We establish convergence to (approximate) equilibria in two-agent zero-sum and multi-agent identical-interest stochastic games over an infinite horizon. This provides one of the first payoff-based and fully decentralized learning algorithms with theoretical guarantees in both settings. Empirical results further validate the robustness and effectiveness of the proposed approach across both classes of games.

cs.LG↗

arXiv2025

Distributionally Robust Multi-Agent Reinforcement Learning for Intelligent Traffic Control

Learning-based traffic signal control is typically optimized for average performance under a few nominal demand patterns, which can result in poor behavior under atypical traffic conditions. To address this, we develop a distributionally robust multi-agent reinforcement learning framework for signal control on a 3x3 urban grid calibrated from a contiguous 3x3 subarea of central Athens covered by the pNEUMA trajectory dataset (Barmpounakis and Geroliminis, 2020). Our approach proceeds in three stages. First, we train a baseline multi-agent RL controller in which each intersection is governed by a proximal policy optimization agent with discrete signal phases, using a centralized training, decentralized execution paradigm. Second, to capture demand uncertainty, we construct eight heterogeneous origin-destination-based traffic scenarios-one directly derived from pNEUMA and seven synthetically generated-to span a wide range of spatial and temporal demand patterns. Over this scenario set, we train a contextual-bandit worst-case estimator that assigns mixture weights to estimate adversarial demand distributions conditioned on context. Finally, without modifying the controller architecture, we fine-tune the baseline multi-agent reinforcement learning agents under these estimated worst-case mixtures to obtain a distributionally robust multi-agent reinforcement learning controller. Across all eight scenarios, as well as on an unseen validation network based on the Sioux Falls configuration, the distributionally robust multi-agent reinforcement learning controller consistently reduces horizon-averaged queues and increases average speeds relative to the baseline, achieving up to 51% shorter queues and 38% higher speeds on the worst-performing scenarios.

eess.SY↗

arXiv2025

Omniscient Attacker in Stochastic Security Games with Interdependent Nodes

The adoption of reinforcement learning for critical infrastructure defense introduces a vulnerability where sophisticated attackers can strategically exploit the defense algorithm's learning dynamics. While prior work addresses this vulnerability in the context of repeated normal-form games, its extension to the stochastic games remains an open research gap. We close this gap by examining stochastic security games between an RL defender and an omniscient attacker, utilizing a tractable linear influence network model. To overcome the structural limitations of prior methods, we propose and apply neuro-dynamic programming. Our experimental results demonstrate that the omniscient attacker can significantly outperform a naive defender, highlighting the critical vulnerability introduced by the learning dynamics and the effectiveness of the proposed strategy.

cs.GT↗

arXiv2025

Achieving Pareto Optimality in Games via Single-bit Feedback

Efficient coordination in multi-agent systems often incurs high communication overhead or slow convergence rates, making scalable welfare optimization difficult. We propose Single-Bit Coordination Dynamics for Pareto-Efficient Outcomes (SBC-PE), a decentralized learning algorithm requiring only a single-bit satisfaction signal per agent each round. Despite this extreme efficiency, SBC-PE guarantees convergence to the exact optimal solution in arbitrary finite games. We establish explicit regret bounds, showing expected regret grows only logarithmically with the horizon, i.e., O(log T). Compared with prior payoff-based or bandit-style rules, SBC-PE uniquely combines minimal signaling, general applicability, and finite-time guarantees. These results show scalable welfare optimization is achievable under minimal communication constraints.

cs.GT↗

arXiv2025

Logit-Q Dynamics for Efficient Learning in Stochastic Teams

We present a new family of logit-Q dynamics for efficient learning in stochastic games by combining the log-linear learning (also known as logit dynamics) for the repeated play of normal-form games with Q-learning for unknown Markov decision processes within the auxiliary stage-game framework. In this framework, we view stochastic games as agents repeatedly playing some stage game associated with the current state of the underlying game while the agents' Q-functions determine the payoffs of these stage games. We show that the logit-Q dynamics presented reach (near) efficient equilibrium in stochastic teams with unknown dynamics and quantify the approximation error. We also show the rationality of the logit-Q dynamics against agents following pure stationary strategies and the convergence of the dynamics in stochastic games where the stage-payoffs induce potential games, yet only a single agent controls the state transitions beyond stochastic teams. The key idea is to approximate the dynamics with a fictional scenario where the Q-function estimates are stationary over epochs whose lengths grow at a sufficiently slow rate. We then couple the dynamics in the main and fictional scenarios to show that these two scenarios become more and more similar across epochs due to the vanishing step size and growing epoch lengths.

cs.GT↗

arXiv2024

Team-Fictitious Play for Reaching Team-Nash Equilibrium in Multi-team Games

Multi-team games, prevalent in robotics and resource management, involve team members striving for a joint best response against other teams. Team-Nash equilibrium (TNE) predicts the outcomes of such coordinated interactions. However, can teams of self-interested agents reach TNE? We introduce Team-Fictitious Play (Team-FP), a new variant of fictitious play where agents respond to the last actions of team members and the beliefs formed about other teams with some inertia in action updates. This design is essential in team coordination beyond the classical fictitious play dynamics. We focus on zero-sum potential team games (ZSPTGs) where teams can interact pairwise while the team members do not necessarily have identical payoffs. We show that Team-FP reaches near TNE in ZSPTGs with a quantifiable error bound. We extend Team-FP dynamics to multi-team Markov games for model-based and model-free cases. The convergence analysis tackles the challenge of non-stationarity induced by evolving opponent strategies based on the optimal coupling lemma and stochastic differential inclusion approximation methods. Our work strengthens the foundation for using TNE to predict the behavior of decentralized teams and offers a practical rule for team learning in multi-team environments. We provide extensive simulations of Team-FP dynamics and compare its performance with other widely studied dynamics such as smooth fictitious play and multiplicative weights update. We further explore how different parameters impact the speed of convergence.

cs.GT↗

arXiv2024

Generalized Individual Q-learning for Polymatrix Games with Partial Observations

This paper addresses the challenge of limited observations in non-cooperative multi-agent systems where agents can have partial access to other agents' actions. We present the generalized individual Q-learning dynamics that combine belief-based and payoff-based learning for the networked interconnections of more than two self-interested agents. This approach leverages access to opponents' actions whenever possible, demonstrably achieving a faster (guaranteed) convergence to quantal response equilibrium in multi-agent zero-sum and potential polymatrix games. Notably, the dynamics reduce to the well-studied smoothed fictitious play and individual Q-learning under full and no access to opponent actions, respectively. We further quantify the improvement in convergence rate due to observing opponents' actions through numerical simulations.

cs.GT↗

arXiv2024

Strategizing against Q-learners: A Control-theoretical Approach

In this paper, we explore the susceptibility of the independent Q-learning algorithms (a classical and widely used multi-agent reinforcement learning method) to strategic manipulation of sophisticated opponents in normal-form games played repeatedly. We quantify how much strategically sophisticated agents can exploit naive Q-learners if they know the opponents' Q-learning algorithm. To this end, we formulate the strategic actors' interactions as a stochastic game (whose state encompasses Q-function estimates of the Q-learners) as if the Q-learning algorithms are the underlying dynamical system. We also present a quantization-based approximation scheme to tackle the continuum state space and analyze its performance for two competing strategic actors and a single strategic actor both analytically and numerically.

cs.GT↗

arXiv2024

Finite-horizon Approximations and Episodic Equilibrium for Stochastic Games

This paper proposes a finite-horizon approximation scheme and introduces episodic equilibrium as a solution concept for stochastic games (SGs), where agents strategize based on the current state and episode stage. The paper also establishes an upper bound on the approximation error that decays with the episode length for both discounted and time-averaged utilities. This approach bridges the gap in the analysis of finite and infinite-horizon SGs, and provides a unifying framework to address time-averaged and discounted utilities. To show the effectiveness of the scheme, the paper presents episodic, decentralized (i.e., payoff-based), and model-free learning dynamics proven to reach (near) episodic equilibrium in broad classes of SGs, including zero-sum, identical-interest and specific general-sum SGs with switching controllers for both time-averaged and discounted utilities.

cs.GT↗

arXiv2023

Convergence of Heterogeneous Learning Dynamics in Zero-sum Stochastic Games

This paper presents new families of algorithms for the repeated play of two-agent (near) zero-sum games and two-agent zero-sum stochastic games. For example, the family includes fictitious play and its variants as members. Commonly, the algorithms in this family are all uncoupled, rational, and convergent even in heterogeneous cases, e.g., where the dynamics may differ in terms of learning rates, full, none or temporal access to opponent actions, and model-based vs model-free learning. The convergence of heterogeneous dynamics is of practical interest especially in competitive environments since agents may have no means or interests in following the same dynamic with the same parameters. We prove that any mixture of such asymmetries does not impact the algorithms' convergence to equilibrium (or near equilibrium if there is experimentation) in zero-sum games with repeated play and in zero-sum (irreducible) stochastic games with sufficiently small discount factors.

cs.GT↗

arXiv2023

On the Global Convergence of Stochastic Fictitious Play in Stochastic Games with Turn-based Controllers

This paper presents a learning dynamic with almost sure convergence guarantee for any stochastic game with turn-based controllers (on state transitions) as long as stage-payoffs induce a zero-sum or identical-interest game. Stage-payoffs for different states can even have different structures, e.g., by summing to zero in some states and being identical in others. The dynamics presented combines the classical stochastic fictitious play with value iteration for stochastic games. There are two key properties: (i) players play finite horizon stochastic games with increasing lengths within the underlying infinite-horizon stochastic game, and (ii) the turn-based controllers ensure that the auxiliary stage-games (induced from the continuation payoff estimated) are strategically equivalent to zero-sum or identical-interest games.

cs.GT↗

arXiv2023

Episodic Logit-Q Dynamics for Efficient Learning in Stochastic Teams

We present new learning dynamics combining (independent) log-linear learning and value iteration for stochastic games within the auxiliary stage game framework. The dynamics presented provably attain the efficient equilibrium (also known as optimal equilibrium) in identical-interest stochastic games, beyond the recent concentration of progress on provable convergence to some (possibly inefficient) equilibrium. The dynamics are also independent in the sense that agents take actions consistent with their local viewpoint to a reasonable extent rather than seeking equilibrium. These aspects can be of practical interest in the control applications of intelligent and autonomous systems. The key challenges are the convergence to an inefficient equilibrium and the non-stationarity of the environment from a single agent's viewpoint due to the adaptation of others. The log-linear update plays an important role in addressing the former. We address the latter through the play-in-episodes scheme in which the agents update their Q-function estimates only at the end of the episodes.

cs.GT↗

arXiv2023

Episodic Logit-Q Dynamics for Efficient Learning in Stochastic Teams

cs.GT↗

arXiv2022

Fictitious play in zero-sum stochastic games

We present a novel variant of fictitious play dynamics combining classical fictitious play with Q-learning for stochastic games and analyze its convergence properties in two-player zero-sum stochastic games. Our dynamics involves players forming beliefs on the opponent strategy and their own continuation payoff (Q-function), and playing a greedy best response by using the estimated continuation payoffs. Players update their beliefs from observations of opponent actions. A key property of the learning dynamics is that update of the beliefs on Q-functions occurs at a slower timescale than update of the beliefs on strategies. We show both in the model-based and model-free cases (without knowledge of player payoff functions and state transition probabilities), the beliefs on strategies converge to a stationary mixed Nash equilibrium of the zero-sum stochastic game.

cs.GT↗