Relative overgeneralization (RO) is a pathology that can arise in cooperative multi-agent tasks when the optimal joint action's utility falls below that of a sub-optimal joint action. RO can cause the agents to get stuck into local optima or fail to solve cooperative tasks requiring significant coordination between agents within a given timestep. In this work, we empirically find that, in multi-agent reinforcement learning (MARL), both value-based and policy gradient MARL algorithms can suffer from RO and fail to learn effective coordination policies. To better overcome RO, we propose a novel approach called curriculum learning for relative overgeneralization (CURO). To solve a target task that exhibits strong RO, in CURO, we first fine-tune the reward function of the target task to generate source tasks to train the agent. Then, to effectively transfer the knowledge acquired in one task to the next, we use a transfer learning method that combines value function transfer with buffer transfer, which enables more efficient exploration in the target task. CURO is general and can be applied to both value-based and policy gradient MARL methods. We demonstrate that, when applied to QMIX, HAPPO, and HATRPO, CURO can successfully overcome severe RO, achieve improved performance, and outperform baseline methods in a variety of challenging cooperative multi-agent tasks.
Various methods for Multi-Agent Reinforcement Learning (MARL) have been developed with the assumption that agents' policies are based on accurate state information. However, policies learned through Deep Reinforcement Learning (DRL) are susceptible to adversarial state perturbation attacks. In this work, we propose a State-Adversarial Markov Game (SAMG) and make the first attempt to investigate different solution concepts of MARL under state uncertainties. Our analysis shows that the commonly used solution concepts of optimal agent policy and robust Nash equilibrium do not always exist in SAMGs. To circumvent this difficulty, we consider a new solution concept called robust agent policy, where agents aim to maximize the worst-case expected state value. We prove the existence of robust agent policy for finite state and finite action SAMGs. Additionally, we propose a Robust Multi-Agent Adversarial Actor-Critic (RMA3C) algorithm to learn robust policies for MARL agents under state uncertainties. Our experiments demonstrate that our algorithm outperforms existing methods when faced with state perturbations and greatly improves the robustness of MARL policies. Our code is public on https://songyanghan.github.io/what_is_solution/.
Communication is supposed to improve multi-agent collaboration and overall performance in cooperative Multi-agent reinforcement learning (MARL). However, such improvements are prevalently limited in practice since most existing communication schemes ignore communication overheads (e.g., communication delays). In this paper, we demonstrate that ignoring communication delays has detrimental effects on collaborations, especially in delay-sensitive tasks such as autonomous driving. To mitigate this impact, we design a delay-aware multi-agent communication model (DACOM) to adapt communication to delays. Specifically, DACOM introduces a component, TimeNet, that is responsible for adjusting the waiting time of an agent to receive messages from other agents such that the uncertainty associated with delay can be addressed. Our experiments reveal that DACOM has a non-negligible performance improvement over other mechanisms by making a better trade-off between the benefits of communication and the costs of waiting for messages.
Decentralized bilevel optimization has received increasing attention recently due to its foundational role in many emerging multi-agent learning paradigms (e.g., multi-agent meta-learning and multi-agent reinforcement learning) over peer-to-peer edge networks. However, to work with the limited computation and communication capabilities of edge networks, a major challenge in developing decentralized bilevel optimization techniques is to lower sample and communication complexities. This motivates us to develop a new decentralized bilevel optimization called DIAMOND (decentralized single-timescale stochastic approximation with momentum and gradient-tracking). The contributions of this paper are as follows: i) our DIAMOND algorithm adopts a single-loop structure rather than following the natural double-loop structure of bilevel optimization, which offers low computation and implementation complexity; ii) compared to existing approaches, the DIAMOND algorithm does not require any full gradient evaluations, which further reduces both sample and computational complexities; iii) through a careful integration of momentum information and gradient tracking techniques, we show that the DIAMOND algorithm enjoys $\mathcal{O}(\epsilon^{-3/2})$ in sample and communication complexities for achieving an $\epsilon$-stationary solution, both of which are independent of the dataset sizes and significantly outperform existing works. Extensive experiments also verify our theoretical findings.
A critical challenge in multi-agent reinforcement learning(MARL) is for multiple agents to efficiently accomplish complex, long-horizon tasks. The agents often have difficulties in cooperating on common goals, dividing complex tasks, and planning through several stages to make progress. We propose to address these challenges by guiding agents with programs designed for parallelization, since programs as a representation contain rich structural and semantic information, and are widely used as abstractions for long-horizon tasks. Specifically, we introduce Efficient Multi-Agent Reinforcement Learning with Parallel Program Guidance(E-MAPP), a novel framework that leverages parallel programs to guide multiple agents to efficiently accomplish goals that require planning over $10+$ stages. E-MAPP integrates the structural information from a parallel program, promotes the cooperative behaviors grounded in program semantics, and improves the time efficiency via a task allocator. We conduct extensive experiments on a series of challenging, long-horizon cooperative tasks in the Overcooked environment. Results show that E-MAPP outperforms strong baselines in terms of the completion rate, time efficiency, and zero-shot generalization ability by a large margin.
Commonly adopted in the manufacturing and aerospace sectors, digital twin (DT) platforms are increasingly seen as a promising paradigm to control, monitor, and analyze software-based, "open", communication systems. Notably, DT platforms provide a sandbox in which to test artificial intelligence (AI) solutions for communication systems, potentially reducing the need to collect data and test algorithms in the field, i.e., on the physical twin (PT). A key challenge in the deployment of DT systems is to ensure that virtual control optimization, monitoring, and analysis at the DT are safe and reliable, avoiding incorrect decisions caused by "model exploitation". To address this challenge, this paper presents a general Bayesian framework with the aim of quantifying and accounting for model uncertainty at the DT that is caused by limitations in the amount and quality of data available at the DT from the PT. In the proposed framework, the DT builds a Bayesian model of the communication system, which is leveraged to enable core DT functionalities such as control via multi-agent reinforcement learning (MARL), monitoring of the PT for anomaly detection, prediction, data-collection optimization, and counterfactual analysis. To exemplify the application of the proposed framework, we specifically investigate a case-study system encompassing multiple sensing devices that report to a common receiver. Experimental results validate the effectiveness of the proposed Bayesian framework as compared to standard frequentist model-based solutions.
We study a multi-agent reinforcement learning (MARL) problem where the agents interact over a given network. The goal of the agents is to cooperatively maximize the average of their entropy-regularized long-term rewards. To overcome the curse of dimensionality and to reduce communication, we propose a Localized Policy Iteration (LPI) algorithm that provably learns a near-globally-optimal policy using only local information. In particular, we show that, despite restricting each agent's attention to only its $\kappa$-hop neighborhood, the agents are able to learn a policy with an optimality gap that decays polynomially in $\kappa$. In addition, we show the finite-sample convergence of LPI to the global optimal policy, which explicitly captures the trade-off between optimality and computational complexity in choosing $\kappa$. Numerical simulations demonstrate the effectiveness of LPI.
We develop a wall model for large-eddy simulation (LES) that takes into account various pressure-gradient effects using multi-agent reinforcement learning (MARL). The model is trained using low-Reynolds-number flow over periodic hills with agents distributed on the wall along the computational grid points. The model utilizes a wall eddy-viscosity formulation as the boundary condition, which is shown to provide better predictions of the mean velocity field, rather than the typical wall-shear stress formulation. Each agent receives states based on local instantaneous flow quantities at an off-wall location, computes a reward based on the estimated wall-shear stress, and provides an action to update the wall eddy viscosity at each time step. The trained wall model is validated in wall-modeled LES (WMLES) of flow over periodic hills at higher Reynolds numbers, and the results show the effectiveness of the model on flow with pressure gradients. The analysis of the trained model indicates that the model is capable of distinguishing between the various pressure gradient regimes present in the flow.
Multi-agent reinforcement learning (MARL) suffers from the non-stationarity problem, which is the ever-changing targets at every iteration when multiple agents update their policies at the same time. Starting from first principle, in this paper, we manage to solve the non-stationarity problem by proposing bidirectional action-dependent Q-learning (ACE). Central to the development of ACE is the sequential decision-making process wherein only one agent is allowed to take action at one time. Within this process, each agent maximizes its value function given the actions taken by the preceding agents at the inference stage. In the learning phase, each agent minimizes the TD error that is dependent on how the subsequent agents have reacted to their chosen action. Given the design of bidirectional dependency, ACE effectively turns a multiagent MDP into a single-agent MDP. We implement the ACE framework by identifying the proper network representation to formulate the action dependency, so that the sequential decision process is computed implicitly in one forward pass. To validate ACE, we compare it with strong baselines on two MARL benchmarks. Empirical experiments demonstrate that ACE outperforms the state-of-the-art algorithms on Google Research Football and StarCraft Multi-Agent Challenge by a large margin. In particular, on SMAC tasks, ACE achieves 100% success rate on almost all the hard and super-hard maps. We further study extensive research problems regarding ACE, including extension, generalization, and practicability. Code is made available to facilitate further research.
Safe and efficient co-planning of multiple robots in pedestrian participation environments is promising for applications. In this work, a novel multi-robot social-aware efficient cooperative planner that on the basis of off-policy multi-agent reinforcement learning (MARL) under partial dimension-varying observation and imperfect perception conditions is proposed. We adopt temporal-spatial graph (TSG)-based social encoder to better extract the importance of social relation between each robot and the pedestrians in its field of view (FOV). Also, we introduce K-step lookahead reward setting in multi-robot RL framework to avoid aggressive, intrusive, short-sighted, and unnatural motion decisions generated by robots. Moreover, we improve the traditional centralized critic network with multi-head global attention module to better aggregates local observation information among different robots to guide the process of individual policy update. Finally, multi-group experimental results verify the effectiveness of the proposed cooperative motion planner.
Microprocessor architects are increasingly resorting to domain-specific customization in the quest for high-performance and energy-efficiency. As the systems grow in complexity, fine-tuning architectural parameters across multiple sub-systems (e.g., datapath, memory blocks in different hierarchies, interconnects, compiler optimization, etc.) quickly results in a combinatorial explosion of design space. This makes domain-specific customization an extremely challenging task. Prior work explores using reinforcement learning (RL) and other optimization methods to automatically explore the large design space. However, these methods have traditionally relied on single-agent RL/ML formulations. It is unclear how scalable single-agent formulations are as we increase the complexity of the design space (e.g., full stack System-on-Chip design). Therefore, we propose an alternative formulation that leverages Multi-Agent RL (MARL) to tackle this problem. The key idea behind using MARL is an observation that parameters across different sub-systems are more or less independent, thus allowing a decentralized role assigned to each agent. We test this hypothesis by designing domain-specific DRAM memory controller for several workload traces. Our evaluation shows that the MARL formulation consistently outperforms single-agent RL baselines such as Proximal Policy Optimization and Soft Actor-Critic over different target objectives such as low power and latency. To this end, this work opens the pathway for new and promising research in MARL solutions for hardware architecture search.
We approach autonomous drone-based reforestation with a collaborative multi-agent reinforcement learning (MARL) setup. Agents can communicate as part of a dynamically changing network. We explore collaboration and communication on the back of a high-impact problem. Forests are the main resource to control rising CO2 conditions. Unfortunately, the global forest volume is decreasing at an unprecedented rate. Many areas are too large and hard to traverse to plant new trees. To efficiently cover as much area as possible, here we propose a Graph Neural Network (GNN) based communication mechanism that enables collaboration. Agents can share location information on areas needing reforestation, which increases viewed area and planted tree count. We compare our proposed communication mechanism with a multi-agent baseline without the ability to communicate. Results show how communication enables collaboration and increases collective performance, planting precision and the risk-taking propensity of individual agents.
Offline multi-agent reinforcement learning (MARL) aims to learn effective multi-agent policies from pre-collected datasets, which is an important step toward the deployment of multi-agent systems in real-world applications. However, in practice, each individual behavior policy that generates multi-agent joint trajectories usually has a different level of how well it performs. e.g., an agent is a random policy while other agents are medium policies. In the cooperative game with global reward, one agent learned by existing offline MARL often inherits this random policy, jeopardizing the performance of the entire team. In this paper, we investigate offline MARL with explicit consideration on the diversity of agent-wise trajectories and propose a novel framework called Shared Individual Trajectories (SIT) to address this problem. Specifically, an attention-based reward decomposition network assigns the credit to each agent through a differentiable key-value memory mechanism in an offline manner. These decomposed credits are then used to reconstruct the joint offline datasets into prioritized experience replay with individual trajectories, thereafter agents can share their good trajectories and conservatively train their policies with a graph attention network (GAT) based critic. We evaluate our method in both discrete control (i.e., StarCraft II and multi-agent particle environment) and continuous control (i.e, multi-agent mujoco). The results indicate that our method achieves significantly better results in complex and mixed offline multi-agent datasets, especially when the difference of data quality between individual trajectories is large.
Quantum machine learning (QML) has received a lot of attention according to its light training parameter numbers and speeds; and the advances of QML lead to active research on quantum multi-agent reinforcement learning (QMARL). Existing classical multi-agent reinforcement learning (MARL) features non-stationarity and uncertain properties. Therefore, this paper presents a simulation software framework for novel QMARL to control autonomous multi-drones, i.e., quantum multi-drone reinforcement learning. Our proposed framework accomplishes reasonable reward convergence and service quality performance with fewer trainable parameters. Furthermore, it shows more stable training results. Lastly, our proposed software allows us to analyze the training process and results.
Value Decomposition (VD) aims to deduce the contributions of agents for decentralized policies in the presence of only global rewards, and has recently emerged as a powerful credit assignment paradigm for tackling cooperative Multi-Agent Reinforcement Learning (MARL) problems. One of the main challenges in VD is to promote diverse behaviors among agents, while existing methods directly encourage the diversity of learned agent networks with various strategies. However, we argue that these dedicated designs for agent networks are still limited by the indistinguishable VD network, leading to homogeneous agent behaviors and thus downgrading the cooperation capability. In this paper, we propose a novel Contrastive Identity-Aware learning (CIA) method, explicitly boosting the credit-level distinguishability of the VD network to break the bottleneck of multi-agent diversity. Specifically, our approach leverages contrastive learning to maximize the mutual information between the temporal credits and identity representations of different agents, encouraging the full expressiveness of credit assignment and further the emergence of individualities. The algorithm implementation of the proposed CIA module is simple yet effective that can be readily incorporated into various VD architectures. Experiments on the SMAC benchmarks and across different VD backbones demonstrate that the proposed method yields results superior to the state-of-the-art counterparts. Our code is available at https://github.com/liushunyu/CIA.
Due to the representation limitation of the joint Q value function, multi-agent reinforcement learning methods with linear value decomposition (LVD) or monotonic value decomposition (MVD) suffer from relative overgeneralization. As a result, they can not ensure optimal consistency (i.e., the correspondence between individual greedy actions and the maximal true Q value). In this paper, we derive the expression of the joint Q value function of LVD and MVD. According to the expression, we draw a transition diagram, where each self-transition node (STN) is a possible convergence. To ensure optimal consistency, the optimal node is required to be the unique STN. Therefore, we propose the greedy-based value representation (GVR), which turns the optimal node into an STN via inferior target shaping and further eliminates the non-optimal STNs via superior experience replay. In addition, GVR achieves an adaptive trade-off between optimality and stability. Our method outperforms state-of-the-art baselines in experiments on various benchmarks. Theoretical proofs and empirical results on matrix games demonstrate that GVR ensures optimal consistency under sufficient exploration.
Oversubscription is a common practice for improving cloud resource utilization. It allows the cloud service provider to sell more resources than the physical limit, assuming not all users would fully utilize the resources simultaneously. However, how to design an oversubscription policy that improves utilization while satisfying the some safety constraints remains an open problem. Existing methods and industrial practices are over-conservative, ignoring the coordination of diverse resource usage patterns and probabilistic constraints. To address these two limitations, this paper formulates the oversubscription for cloud as a chance-constrained optimization problem and propose an effective Chance Constrained Multi-Agent Reinforcement Learning (C2MARL) method to solve this problem. Specifically, C2MARL reduces the number of constraints by considering their upper bounds and leverages a multi-agent reinforcement learning paradigm to learn a safe and optimal coordination policy. We evaluate our C2MARL on an internal cloud platform and public cloud datasets. Experiments show that our C2MARL outperforms existing methods in improving utilization ($20\%\sim 86\%$) under different levels of safety constraints.
Distributed access control is a crucial component for massive machine type communication (mMTC). In this communication scenario, centralized resource allocation is not scalable because resource configurations have to be sent frequently from the base station to a massive number of devices. We investigate distributed reinforcement learning for resource selection without relying on centralized control. Another important feature of mMTC is the sporadic and dynamic change of traffic. Existing studies on distributed access control assume that traffic load is static or they are able to gradually adapt to the dynamic traffic. We minimize the adaptation period by training TinyQMIX, which is a lightweight multi-agent deep reinforcement learning model, to learn a distributed wireless resource selection policy under various traffic patterns before deployment. Therefore, the trained agents are able to quickly adapt to dynamic traffic and provide low access delay. Numerical results are presented to support our claims.
The energy transition potentially poses an existential risk for major international oil companies (IOCs) if they fail to adapt to low-carbon business models. Projections of energy futures, however, are met with diverging assumptions on its scale and pace, causing disagreement among IOC decision-makers and their stakeholders over what the business model of an incumbent fossil fuel company should be. In this work, we used deep multi-agent reinforcement learning to solve an energy systems wargame wherein players simulate IOC decision-making, including hydrocarbon and low-carbon investments decisions, dividend policies, and capital structure measures, through an uncertain energy transition to explore critical and non-linear governance questions, from leveraged transitions to reserve replacements. Adversarial play facilitated by state-of-the-art algorithms revealed decision-making strategies robust to energy transition uncertainty and against multiple IOCs. In all games, robust strategies emerged in the form of low-carbon business models as a result of early transition-oriented movement. IOCs adopting such strategies outperformed business-as-usual and delayed transition strategies regardless of hydrocarbon demand projections. In addition to maximizing value, these strategies benefit greater society by contributing substantial amounts of capital necessary to accelerate the global low-carbon energy transition. Our findings point towards the need for lenders and investors to effectively mobilize transition-oriented finance and engage with IOCs to ensure responsible reallocation of capital towards low-carbon business models that would enable the emergence of fossil fuel incumbents as future low-carbon leaders.
Traditional multi-agent reinforcement learning (MARL) algorithms, such as independent Q-learning, struggle when presented with partially observable scenarios, and where agents are required to develop delicate action sequences. This is often the result of the reward for a good action only being available after other agents have taken theirs, and these actions are not credited accordingly. Recurrent neural networks have proven to be a viable solution strategy for solving these types of problems, resulting in significant performance increase when compared to other methods. In this paper, we explore a different approach and focus on the experiences used to update the action-value functions of each agent. We introduce the concept of credit-cognisant rewards (CCRs), which allows an agent to perceive the effect its actions had on the environment as well as on its co-agents. We show that by manipulating these experiences and constructing the reward contained within them to include the rewards received by all the agents within the same action sequence, we are able to improve significantly on the performance of independent deep Q-learning as well as deep recurrent Q-learning. We evaluate and test the performance of CCRs when applied to deep reinforcement learning techniques at the hands of a simplified version of the popular card game Hanabi.