On-demand ride-sharing platforms face the fundamental challenge of dynamically bundling passengers with diverse origins and destinations and matching them with vehicles in real time, all under significant uncertainty. Recently, MARL has emerged as a promising solution for this problem, leveraging decentralized learning to address the curse of dimensionality caused by the large number of agents in the ride-hailing market and the resulting expansive state and action spaces. However, conventional MARL-based ride-sharing approaches heavily rely on the accurate estimation of Q-values or V-values, which becomes problematic in large-scale, highly uncertain environments. Specifically, most of these approaches adopt an independent paradigm, exacerbating this issue, as each agent treats others as part of the environment, leading to unstable training and substantial estimation bias in value functions. To address these challenges, we propose two novel alternative methods that bypass value function estimation. First, we adapt GRPO to ride-sharing, replacing the PPO baseline with the group average reward to eliminate critic estimation errors and reduce training bias. Second, inspired by GRPO's full utilization of group reward information, we customize the PPO framework for ride-sharing platforms and show that, under a homogeneous fleet, the optimal policy can be trained using only one-step rewards - a method we term One-Step Policy Optimization (OSPO). Experiments on a real-world Manhattan ride-hailing dataset demonstrate that both GRPO and OSPO achieve superior performance across most scenarios, efficiently optimizing pickup times and the number of served orders using simple MLP networks.
Recent advances in reinforcement learning have shown that language models can develop sophisticated reasoning through training on tasks with verifiable rewards, but these approaches depend on human-curated problem-answer pairs and domain-specific reward engineering. We introduce SPIRAL, a self-play framework where models learn by playing multi-turn, zero-sum games against continuously improving versions of themselves, eliminating the need for human supervision. Through self-play, SPIRAL generates an infinite curriculum of progressively challenging problems as models must constantly adapt to stronger opponents. To enable this self-play training at scale, We implement a fully online, multi-turn, multi-agent reinforcement learning system for LLMs and propose role-conditioned advantage estimation (RAE) to stabilize multi-agent training. Using SPIRAL, self-play on zero-sum games produces reasoning capabilities that transfer broadly. Training Qwen3-4B-Base on Kuhn Poker alone achieves 8.6% improvement on math and 8.4% on general reasoning, outperforming SFT on 25,000 expert game trajectories. Analysis reveals that this transfer occurs through three cognitive patterns: systematic decomposition, expected value calculation, and case-by-case analysis. Multi-game training (TicTacToe, Kuhn Poker, Simple Negotiation) further enhances performance as each game develops distinct reasoning strengths. Applying SPIRAL to a strong reasoning model (DeepSeek-R1-Distill-Qwen-7B) can still lead to 2.0% average improvement. These results demonstrate that zero-sum games naturally develop transferable reasoning capabilities, highlighting a promising direction for autonomous reasoning development.
We present the Multi-Agent Transformer World Model (MATWM), a novel transformer-based world model designed for multi-agent reinforcement learning in both vector- and image-based environments. MATWM combines a decentralized imagination framework with a semi-centralized critic and a teammate prediction module, enabling agents to model and anticipate the behavior of others under partial observability. To address non-stationarity, we incorporate a prioritized replay mechanism that trains the world model on recent experiences, allowing it to adapt to agents' evolving policies. We evaluated MATWM on a broad suite of benchmarks, including the StarCraft Multi-Agent Challenge, PettingZoo, and MeltingPot. MATWM achieves state-of-the-art performance, outperforming both model-free and prior world model approaches, while demonstrating strong sample efficiency, achieving near-optimal performance in as few as 50K environment interactions. Ablation studies confirm the impact of each component, with substantial gains in coordination-heavy tasks.
Adapting a single agent to a new multi-agent system brings challenges, necessitating adjustments across various tasks, environments, and interactions with unknown teammates and opponents. Addressing this challenge is highly complex, and researchers have proposed two simplified scenarios, Multi-agent reinforcement learning for zero-shot learning and Ad-Hoc Teamwork. Building on these foundations, we propose a more comprehensive setting, Agent Collaborative-Competitive Adaptation (ACCA), which evaluates an agent to generalize across diverse scenarios, tasks, and interactions with both unfamiliar opponents and teammates. In ACCA, agents adjust to task and environmental changes, collaborate with unseen teammates, and compete against unknown opponents. We introduce a new modeling approach, Multi-Retrieval and Dynamic Generation (MRDG), that effectively models both teammates and opponents using their behavioral trajectories. This method incorporates a positional encoder for varying team sizes and a hypernetwork module to boost agents' learning and adaptive capabilities. Additionally, a viewpoint alignment module harmonizes the observational perspectives of retrieved teammates and opponents with the learning agent. Extensive tests in benchmark scenarios like SMAC, Overcooked-AI, and Melting Pot show that MRDG significantly improves robust collaboration and competition with unseen teammates and opponents, surpassing established baselines. Our code is available at: https://github.com/vcis-wangchenxu/MRDG.git
As Large Language Models (LLMs) gain agentic abilities, they will have to navigate complex multi-agent scenarios, interacting with human users and other agents in cooperative and competitive settings. This will require new reasoning skills, chief amongst them being theory of mind (ToM), or the ability to reason about the "mental" states of other agents. However, ToM and other multi-agent abilities in LLMs are poorly understood, since existing benchmarks suffer from narrow scope, data leakage, saturation, and lack of interactivity. We thus propose Decrypto, a game-based benchmark for multi-agent reasoning and ToM drawing inspiration from cognitive science, computational pragmatics and multi-agent reinforcement learning. It is designed to be as easy as possible in all other dimensions, eliminating confounding factors commonly found in other benchmarks. To our knowledge, it is also the first platform for designing interactive ToM experiments. We validate the benchmark design through comprehensive empirical evaluations of frontier LLMs, robustness studies, and human-AI cross-play experiments. We find that LLM game-playing abilities lag behind humans and simple word-embedding baselines. We then create variants of two classic cognitive science experiments within Decrypto to evaluate three key ToM abilities. Surprisingly, we find that state-of-the-art reasoning models are significantly worse at those tasks than their older counterparts. This demonstrates that Decrypto addresses a crucial gap in current reasoning and ToM evaluations, and paves the path towards better artificial agents.
This paper investigates multi-agent reinforcement learning (MARL) in a partially observable, cooperative-competitive combat environment known as LAG. We describe the environment's setup, including agent actions, hierarchical controls, and reward design across different combat modes such as No Weapon and ShootMissile. Two representative algorithms are evaluated: HAPPO, an on-policy hierarchical variant of PPO, and HASAC, an off-policy method based on soft actor-critic. We analyze their training stability, reward progression, and inter-agent coordination capabilities. Experimental results show that HASAC performs well in simpler coordination tasks without weapons, while HAPPO demonstrates stronger adaptability in more dynamic and expressive scenarios involving missile combat. These findings provide insights into the trade-offs between on-policy and off-policy methods in multi-agent settings.
Multi-agent reinforcement learning (MARL) has achieved strong performance in cooperative adversarial tasks. However, most existing methods typically train agents against fixed opponent strategies and rely on such meta-static difficulty conditions, which limits their adaptability to changing environments and often leads to suboptimal policies. Inspired by the success of curriculum learning (CL) in supervised tasks, we propose a dynamic CL framework for MARL that employs an self-adaptive difficulty adjustment mechanism. This mechanism continuously modulates opponent strength based on real-time agent training performance, allowing agents to progressively learn from easier to more challenging scenarios. However, the dynamic nature of CL introduces instability due to nonstationary environments and sparse global rewards. To address this challenge, we develop a Counterfactual Group Relative Policy Advantage (CGRPA), which is tightly coupled with the curriculum by providing intrinsic credit signals that reflect each agent's impact under evolving task demands. CGRPA constructs a counterfactual advantage function that isolates individual contributions within group behavior, facilitating more reliable policy updates throughout the curriculum. CGRPA evaluates each agent's contribution through constructing counterfactual action advantage function, providing intrinsic rewards that enhance credit assignment and stabilize learning under non-stationary conditions. Extensive experiments demonstrate that our method improves both training stability and final performance, achieving competitive results against state-of-the-art methods. The code is available at https://github.com/NICE-HKU/CL2MARL-SMAC.
Conventional language model (LM) safety alignment relies on a reactive, disjoint procedure: attackers exploit a static model, followed by defensive fine-tuning to patch exposed vulnerabilities. This sequential approach creates a mismatch -- attackers overfit to obsolete defenses, while defenders perpetually lag behind emerging threats. To address this, we propose Self-RedTeam, an online self-play reinforcement learning algorithm where an attacker and defender agent co-evolve through continuous interaction. We cast safety alignment as a two-player zero-sum game, where a single model alternates between attacker and defender roles -- generating adversarial prompts and safeguarding against them -- while a reward LM adjudicates outcomes. This enables dynamic co-adaptation. Grounded in the game-theoretic framework of zero-sum games, we establish a theoretical safety guarantee which motivates the design of our method: if self-play converges to a Nash Equilibrium, the defender will reliably produce safe responses to any adversarial input. Empirically, Self-RedTeam uncovers more diverse attacks (+21.8% SBERT) compared to attackers trained against static defenders and achieves higher robustness on safety benchmarks (e.g., +65.5% on WildJailBreak) than defenders trained against static attackers. We further propose hidden Chain-of-Thought, allowing agents to plan privately, which boosts adversarial diversity and reduces over-refusals. Our results motivate a shift from reactive patching to proactive co-evolution in LM safety training, enabling scalable, autonomous, and robust self-improvement of LMs via multi-agent reinforcement learning (MARL).
We introduce Sorrel (https://github.com/social-ai-uoft/sorrel), a simple Python interface for generating and testing new multi-agent reinforcement learning environments. This interface places a high degree of emphasis on simplicity and accessibility, and uses a more psychologically intuitive structure for the basic agent-environment loop, making it a useful tool for social scientists to investigate how learning and social interaction leads to the development and change of group dynamics. In this short paper, we outline the basic design philosophy and features of Sorrel.
Gene selection in high-dimensional genomic data is essential for understanding disease mechanisms and improving therapeutic outcomes. Traditional feature selection methods effectively identify predictive genes but often ignore complex biological pathways and regulatory networks, leading to unstable and biologically irrelevant signatures. Prior approaches, such as Lasso-based methods and statistical filtering, either focus solely on individual gene-outcome associations or fail to capture pathway-level interactions, presenting a key challenge: how to integrate biological pathway knowledge while maintaining statistical rigor in gene selection? To address this gap, we propose a novel two-stage framework that integrates statistical selection with biological pathway knowledge using multi-agent reinforcement learning (MARL). First, we introduce a pathway-guided pre-filtering strategy that leverages multiple statistical methods alongside KEGG pathway information for initial dimensionality reduction. Next, for refined selection, we model genes as collaborative agents in a MARL framework, where each agent optimizes both predictive power and biological relevance. Our framework incorporates pathway knowledge through Graph Neural Network-based state representations, a reward mechanism combining prediction performance with gene centrality and pathway coverage, and collaborative learning strategies using shared memory and a centralized critic component. Extensive experiments on multiple gene expression datasets demonstrate that our approach significantly improves both prediction accuracy and biological interpretability compared to traditional methods.
World models have recently attracted growing interest in Multi-Agent Reinforcement Learning (MARL) due to their ability to improve sample efficiency for policy learning. However, accurately modeling environments in MARL is challenging due to the exponentially large joint action space and highly uncertain dynamics inherent in multi-agent systems. To address this, we reduce modeling complexity by shifting from jointly modeling the entire state-action transition dynamics to focusing on the state space alone at each timestep through sequential agent modeling. Specifically, our approach enables the model to progressively resolve uncertainty while capturing the structured dependencies among agents, providing a more accurate representation of how agents influence the state. Interestingly, this sequential revelation of agents' actions in a multi-agent system aligns with the reverse process in diffusion models--a class of powerful generative models known for their expressiveness and training stability compared to autoregressive or latent variable models. Leveraging this insight, we develop a flexible and robust world model for MARL using diffusion models. Our method, Diffusion-Inspired Multi-Agent world model (DIMA), achieves state-of-the-art performance across multiple multi-agent control benchmarks, significantly outperforming prior world models in terms of final return and sample efficiency, including MAMuJoCo and Bi-DexHands. DIMA establishes a new paradigm for constructing multi-agent world models, advancing the frontier of MARL research.
Urban traffic congestion, particularly at intersections, significantly impacts travel time, fuel consumption, and emissions. Traditional fixed-time signal control systems often lack the adaptability to manage dynamic traffic patterns effectively. This study explores the application of multi-agent reinforcement learning (MARL) to optimize traffic signal coordination across multiple intersections within a simulated environment. Utilizing Pygame, a simulation was developed to model a network of interconnected intersections with randomly generated vehicle flows to reflect realistic traffic variability. A decentralized MARL controller was implemented, in which each traffic signal operates as an autonomous agent, making decisions based on local observations and information from neighboring agents. Performance was evaluated against a baseline fixed-time controller using metrics such as average vehicle wait time and overall throughput. The MARL approach demonstrated statistically significant improvements, including reduced average waiting times and improved throughput. These findings suggest that MARL-based dynamic control strategies hold substantial promise for improving urban traffic management efficiency. More research is recommended to address scalability and real-world implementation challenges.
Sequential social dilemmas pose a significant challenge in the field of multi-agent reinforcement learning (MARL), requiring environments that accurately reflect the tension between individual and collective interests. Previous benchmarks and environments, such as Melting Pot, provide an evaluation protocol that measures generalization to new social partners in various test scenarios. However, running reinforcement learning algorithms in traditional environments requires substantial computational resources. In this paper, we introduce SocialJax, a suite of sequential social dilemma environments and algorithms implemented in JAX. JAX is a high-performance numerical computing library for Python that enables significant improvements in operational efficiency. Our experiments demonstrate that the SocialJax training pipeline achieves at least 50\texttimes{} speed-up in real-time performance compared to Melting Pot RLlib baselines. Additionally, we validate the effectiveness of baseline algorithms within SocialJax environments. Finally, we use Schelling diagrams to verify the social dilemma properties of these environments, ensuring that they accurately capture the dynamics of social dilemmas.
Multi-agent reinforcement learning (MARL) has emerged as a promising solution for learning complex and scalable coordination behaviors in multi-robot systems. However, established MARL platforms (e.g., SMAC and MPE) lack robotics relevance and hardware deployment, leaving multi-robot learning researchers to develop bespoke environments and hardware testbeds dedicated to the development and evaluation of their individual contributions. The Multi-Agent RL Benchmark and Learning Environment for the Robotarium (MARBLER) is an exciting recent step in providing a standardized robotics-relevant platform for MARL, by bridging the Robotarium testbed with existing MARL software infrastructure. However, MARBLER lacks support for parallelization and GPU/TPU execution, making the platform prohibitively slow compared to modern MARL environments and hindering adoption. We contribute JaxRobotarium, a Jax-powered end-to-end simulation, learning, deployment, and benchmarking platform for the Robotarium. JaxRobotarium enables rapid training and deployment of multi-robot RL (MRRL) policies with realistic robot dynamics and safety constraints, supporting parallelization and hardware acceleration. Our generalizable learning interface integrates easily with SOTA MARL libraries (e.g., JaxMARL). In addition, JaxRobotarium includes eight standardized coordination scenarios, including four novel scenarios that bring established MARL benchmark tasks (e.g., RWARE and Level-Based Foraging) to a robotics setting. We demonstrate that JaxRobotarium retains high simulation fidelity while achieving dramatic speedups over baseline (20x in training and 150x in simulation), and provides an open-access sim-to-real evaluation pipeline through the Robotarium testbed, accelerating and democratizing access to multi-robot learning research and evaluation. Our code is available at https://github.com/GT-STAR-Lab/JaxRobotarium.
Learning to cooperate in distributed partially observable environments with no communication abilities poses significant challenges for multi-agent deep reinforcement learning (MARL). This paper addresses key concerns in this domain, focusing on inferring state representations from individual agent observations and leveraging these representations to enhance agents' exploration and collaborative task execution policies. To this end, we propose a novel state modelling framework for cooperative MARL, where agents infer meaningful belief representations of the non-observable state, with respect to optimizing their own policies, while filtering redundant and less informative joint state information. Building upon this framework, we propose the MARL SMPE algorithm. In SMPE, agents enhance their own policy's discriminative abilities under partial observability, explicitly by incorporating their beliefs into the policy network, and implicitly by adopting an adversarial type of exploration policies which encourages agents to discover novel, high-value states while improving the discriminative abilities of others. Experimentally, we show that SMPE outperforms state-of-the-art MARL algorithms in complex fully cooperative tasks from the MPE, LBF, and RWARE benchmarks.
DBSCAN, a well-known density-based clustering algorithm, has gained widespread popularity and usage due to its effectiveness in identifying clusters of arbitrary shapes and handling noisy data. However, it encounters challenges in producing satisfactory cluster results when confronted with datasets of varying density scales, a common scenario in real-world applications. In this paper, we propose a novel Adaptive and Robust DBSCAN with Multi-agent Reinforcement Learning cluster framework, namely AR-DBSCAN. First, we model the initial dataset as a two-level encoding tree and categorize the data vertices into distinct density partitions according to the information uncertainty determined in the encoding tree. Each partition is then assigned to an agent to find the best clustering parameters without manual assistance. The allocation is density-adaptive, enabling AR-DBSCAN to effectively handle diverse density distributions within the dataset by utilizing distinct agents for different partitions. Second, a multi-agent deep reinforcement learning guided automatic parameter searching process is designed. The process of adjusting the parameter search direction by perceiving the clustering environment is modeled as a Markov decision process. Using a weakly-supervised reward training policy network, each agent adaptively learns the optimal clustering parameters by interacting with the clusters. Third, a recursive search mechanism adaptable to the data's scale is presented, enabling efficient and controlled exploration of large parameter spaces. Extensive experiments are conducted on nine artificial datasets and a real-world dataset. The results of offline and online tasks show that AR-DBSCAN not only improves clustering accuracy by up to 144.1% and 175.3% in the NMI and ARI metrics, respectively, but also is capable of robustly finding dominant parameters.
Multi-Agent Reinforcement Learning (MARL) has become a powerful framework for numerous real-world applications, modeling distributed decision-making and learning from interactions with complex environments. Resource Allocation Optimization (RAO) benefits significantly from MARL's ability to tackle dynamic and decentralized contexts. MARL-based approaches are increasingly applied to RAO challenges across sectors playing pivotal roles to Industry 4.0 developments. This survey provides a comprehensive review of recent MARL algorithms for RAO, encompassing core concepts, classifications, and a structured taxonomy. By outlining the current research landscape and identifying primary challenges and future directions, this survey aims to support researchers and practitioners in leveraging MARL's potential to advance resource allocation solutions.
This paper explores the emergence of language in multi-agent reinforcement learning (MARL) using transformers. Existing methods such as RIAL, DIAL, and CommNet enable agent communication but lack interpretability. We propose Differentiable Inter-Agent Transformers (DIAT), which leverage self-attention to learn symbolic, human-understandable communication protocols. Through experiments, DIAT demonstrates the ability to encode observations into interpretable vocabularies and meaningful embeddings, effectively solving cooperative tasks. These results highlight the potential of DIAT for interpretable communication in complex multi-agent environments.
Traffic congestion remains a significant challenge in modern urban networks. Autonomous driving technologies have emerged as a potential solution. Among traffic control methods, reinforcement learning has shown superior performance over traffic signals in various scenarios. However, prior research has largely focused on small-scale networks or isolated intersections, leaving large-scale mixed traffic control largely unexplored. This study presents the first attempt to use decentralized multi-agent reinforcement learning for large-scale mixed traffic control in which some intersections are managed by traffic signals and others by robot vehicles. Evaluating a real-world network in Colorado Springs, CO, USA with 14 intersections, we measure traffic efficiency via average waiting time of vehicles at intersections and the number of vehicles reaching their destinations within a time window (i.e., throughput). At 80% RV penetration rate, our method reduces waiting time from 6.17 s to 5.09 s and increases throughput from 454 vehicles per 500 seconds to 493 vehicles per 500 seconds, outperforming the baseline of fully signalized intersections. These findings suggest that integrating reinforcement learning-based control large-scale traffic can improve overall efficiency and may inform future urban planning strategies.
Traffic congestion in modern cities is exacerbated by the limitations of traditional fixed-time traffic signal systems, which fail to adapt to dynamic traffic patterns. Adaptive Traffic Signal Control (ATSC) algorithms have emerged as a solution by dynamically adjusting signal timing based on real-time traffic conditions. However, the main limitation of such methods is that they are not transferable to environments under real-world constraints, such as balancing efficiency, minimizing collisions, and ensuring fairness across intersections. In this paper, we view the ATSC problem as a constrained multi-agent reinforcement learning (MARL) problem and propose a novel algorithm named Multi-Agent Proximal Policy Optimization with Lagrange Cost Estimator (MAPPO-LCE) to produce effective traffic signal control policies. Our approach integrates the Lagrange multipliers method to balance rewards and constraints, with a cost estimator for stable adjustment. We also introduce three constraints on the traffic network: GreenTime, GreenSkip, and PhaseSkip, which penalize traffic policies that do not conform to real-world scenarios. Our experimental results on three real-world datasets demonstrate that MAPPO-LCE outperforms three baseline MARL algorithms by across all environments and traffic constraints (improving on MAPPO by 12.60%, IPPO by 10.29%, and QTRAN by 13.10%). Our results show that constrained MARL is a valuable tool for traffic planners to deploy scalable and efficient ATSC methods in real-world traffic networks. We provide code at https://github.com/Asatheesh6561/MAPPO-LCE.