PyTorchKR Papers

최신 논문

총 1,869편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2303.00177

Finite-sample Guarantees for Nash Q-learning with Linear Function Approximation

Nash Q-learning may be considered one of the first and most known algorithms in multi-agent reinforcement learning (MARL) for learning policies that constitute a Nash equilibrium of an underlying general-sum Markov game. Its original proof provided asymptotic guarantees and was for the tabular case. Recently, finite-sample guarantees have been provided using more modern RL techniques for the tabular case. Our work analyzes Nash Q-learning using linear function approximation -- a representation regime introduced when the state space is large or continuous -- and provides finite-sample guarantees that indicate its sample efficiency. We find that the obtained performance nearly matches an existing efficient result for single-agent RL under the same representation and has a polynomial gap when compared to the best-known result for the tabular case.

arXiv:2303.00451

A Variational Approach to Mutual Information-Based Coordination for Multi-Agent Reinforcement Learning

In this paper, we propose a new mutual information framework for multi-agent reinforcement learning to enable multiple agents to learn coordinated behaviors by regularizing the accumulated return with the simultaneous mutual information between multi-agent actions. By introducing a latent variable to induce nonzero mutual information between multi-agent actions and applying a variational bound, we derive a tractable lower bound on the considered MMI-regularized objective function. The derived tractable objective can be interpreted as maximum entropy reinforcement learning combined with uncertainty reduction of other agents actions. Applying policy iteration to maximize the derived lower bound, we propose a practical algorithm named variational maximum mutual information multi-agent actor-critic, which follows centralized learning with decentralized execution. We evaluated VM3-AC for several games requiring coordination, and numerical results show that VM3-AC outperforms other MARL algorithms in multi-agent tasks requiring high-quality coordination.

arXiv:2302.14276

On the Role of Emergent Communication for Social Learning in Multi-Agent Reinforcement Learning

Explicit communication among humans is key to coordinating and learning. Social learning, which uses cues from experts, can greatly benefit from the usage of explicit communication to align heterogeneous policies, reduce sample complexity, and solve partially observable tasks. Emergent communication, a type of explicit communication, studies the creation of an artificial language to encode a high task-utility message directly from data. However, in most cases, emergent communication sends insufficiently compressed messages with little or null information, which also may not be understandable to a third-party listener. This paper proposes an unsupervised method based on the information bottleneck to capture both referential complexity and task-specific utility to adequately explore sparse social communication scenarios in multi-agent reinforcement learning (MARL). We show that our model is able to i) develop a natural-language-inspired lexicon of messages that is independently composed of a set of emergent concepts, which span the observations and intents with minimal bits, ii) develop communication to align the action policies of heterogeneous agents with dissimilar feature models, and iii) learn a communication policy from watching an expert's action policy, which we term `social shadowing'.

arXiv:2302.14094

Combating Uncertainties in Wind and Distributed PV Energy Sources Using Integrated Reinforcement Learning and Time-Series Forecasting

Renewable energy sources, such as wind and solar power, are increasingly being integrated into smart grid systems. However, when compared to traditional energy resources, the unpredictability of renewable energy generation poses significant challenges for both electricity providers and utility companies. Furthermore, the large-scale integration of distributed energy resources (such as PV systems) creates new challenges for energy management in microgrids. To tackle these issues, we propose a novel framework with two objectives: (i) combating uncertainty of renewable energy in smart grid by leveraging time-series forecasting with Long-Short Term Memory (LSTM) solutions, and (ii) establishing distributed and dynamic decision-making framework with multi-agent reinforcement learning using Deep Deterministic Policy Gradient (DDPG) algorithm. The proposed framework considers both objectives concurrently to fully integrate them, while considering both wholesale and retail markets, thereby enabling efficient energy management in the presence of uncertain and distributed renewable energy sources. Through extensive numerical simulations, we demonstrate that the proposed solution significantly improves the profit of load serving entities (LSE) by providing a more accurate wind generation forecast. Furthermore, our results demonstrate that households with PV and battery installations can increase their profits by using intelligent battery charge/discharge actions determined by the DDPG agents.

arXiv:2302.14604공식 코드 · Papers with Code

IQ-Flow: Mechanism Design for Inducing Cooperative Behavior to Self-Interested Agents in Sequential Social Dilemmas

Achieving and maintaining cooperation between agents to accomplish a common objective is one of the central goals of Multi-Agent Reinforcement Learning (MARL). Nevertheless in many real-world scenarios, separately trained and specialized agents are deployed into a shared environment, or the environment requires multiple objectives to be achieved by different coexisting parties. These variations among specialties and objectives are likely to cause mixed motives that eventually result in a social dilemma where all the parties are at a loss. In order to resolve this issue, we propose the Incentive Q-Flow (IQ-Flow) algorithm, which modifies the system's reward setup with an incentive regulator agent such that the cooperative policy also corresponds to the self-interested policy for the agents. Unlike the existing methods that learn to incentivize self-interested agents, IQ-Flow does not make any assumptions about agents' policies or learning algorithms, which enables the generalization of the developed framework to a wider array of applications. IQ-Flow performs an offline evaluation of the optimality of the learned policies using the data provided by other agents to determine cooperative and self-interested policies. Next, IQ-Flow uses meta-gradient learning to estimate how policy evaluation changes according to given incentives and modifies the incentive such that the greedy policy for cooperative objective and self-interested objective yield the same actions. We present the operational characteristics of IQ-Flow in Iterated Matrix Games. We demonstrate that IQ-Flow outperforms the state-of-the-art incentive design algorithm in Escape Room and 2-Player Cleanup environments. We further demonstrate that the pretrained IQ-Flow mechanism significantly outperforms the performance of the shared reward setup in the 2-Player Cleanup environment.

arXiv:2302.14399

Multi-Agent Reinforcement Learning for Pragmatic Communication and Control

The automation of factories and manufacturing processes has been accelerating over the past few years, boosted by the Industry 4.0 paradigm, including diverse scenarios with mobile, flexible agents. Efficient coordination between mobile robots requires reliable wireless transmission in highly dynamic environments, often with strict timing requirements. Goal-oriented communication is a possible solution for this problem: communication decisions should be optimized for the target control task, providing the information that is most relevant to decide which action to take. From the control perspective, networked control design takes the communication impairments into account in its optmization of physical actions. In this work, we propose a joint design that combines goal-oriented communication and networked control into a single optimization model, an extension of a multiagent POMDP which we call Cyber-Physical POMDP (CP-POMDP). The model is flexible enough to represent several swarm and cooperative scenarios, and we illustrate its potential with two simple reference scenarios with a single agent and a set of supporting sensors. Joint training of the communication and control systems can significantly improve the overall performance, particularly if communication is severely constrained, and can even lead to implicit coordination of communication actions.

arXiv:2302.12899

Multi-Agent Reinforcement Learning with Common Policy for Antenna Tilt Optimization

This paper presents a method for optimizing wireless networks by adjusting cell parameters that affect both the performance of the cell being optimized and the surrounding cells. The method uses multiple reinforcement learning agents that share a common policy and take into account information from neighboring cells to determine the state and reward. In order to avoid impairing network performance during the initial stages of learning, agents are pre-trained in an earlier phase of offline learning. During this phase, an initial policy is obtained using feedback from a static network simulator and considering a wide variety of scenarios. Finally, agents can intelligently tune the cell parameters of a test network by suggesting small incremental changes, slowly guiding the network toward an optimal configuration. The agents propose optimal changes using the experience gained with the simulator in the pre-training phase, but they can also continue to learn from current network readings after each change. The results show how the proposed approach significantly improves the performance gains already provided by expert system-based methods when applied to remote antenna tilt optimization. The significant gains of this approach have truly been observed when compared with a similar method in which the state and reward do not incorporate information from neighboring cells.

arXiv:2302.12515

AC2C: Adaptively Controlled Two-Hop Communication for Multi-Agent Reinforcement Learning

Learning communication strategies in cooperative multi-agent reinforcement learning (MARL) has recently attracted intensive attention. Early studies typically assumed a fully-connected communication topology among agents, which induces high communication costs and may not be feasible. Some recent works have developed adaptive communication strategies to reduce communication overhead, but these methods cannot effectively obtain valuable information from agents that are beyond the communication range. In this paper, we consider a realistic communication model where each agent has a limited communication range, and the communication topology dynamically changes. To facilitate effective agent communication, we propose a novel communication protocol called Adaptively Controlled Two-Hop Communication (AC2C). After an initial local communication round, AC2C employs an adaptive two-hop communication strategy to enable long-range information exchange among agents to boost performance, which is implemented by a communication controller. This controller determines whether each agent should ask for two-hop messages and thus helps to reduce the communication overhead during distributed execution. We evaluate AC2C on three cooperative multi-agent tasks, and the experimental results show that it outperforms relevant baselines with lower communication costs.

arXiv:2302.11793공식 코드 · Papers with Code

Revisiting the Gumbel-Softmax in MADDPG

MADDPG is an algorithm in multi-agent reinforcement learning (MARL) that extends the popular single-agent method, DDPG, to multi-agent scenarios. Importantly, DDPG is an algorithm designed for continuous action spaces, where the gradient of the state-action value function exists. For this algorithm to work in discrete action spaces, discrete gradient estimation must be performed. For MADDPG, the Gumbel-Softmax (GS) estimator is used -- a reparameterisation which relaxes a discrete distribution into a similar continuous one. This method, however, is statistically biased, and a recent MARL benchmarking paper suggests that this bias makes MADDPG perform poorly in grid-world situations, where the action space is discrete. Fortunately, many alternatives to the GS exist, boasting a wide range of properties. This paper explores several of these alternatives and integrates them into MADDPG for discrete grid-world scenarios. The corresponding impact on various performance metrics is then measured and analysed. It is found that one of the proposed estimators performs significantly better than the original GS in several tasks, achieving up to 55% higher returns, along with faster convergence.

arXiv:2302.12232

Concept Learning for Interpretable Multi-Agent Reinforcement Learning

Multi-agent robotic systems are increasingly operating in real-world environments in close proximity to humans, yet are largely controlled by policy models with inscrutable deep neural network representations. We introduce a method for incorporating interpretable concepts from a domain expert into models trained through multi-agent reinforcement learning, by requiring the model to first predict such concepts then utilize them for decision making. This allows an expert to both reason about the resulting concept policy models in terms of these high-level concepts at run-time, as well as intervene and correct mispredictions to improve performance. We show that this yields improved interpretability and training stability, with benefits to policy performance and sample efficiency in a simulated and real-world cooperative-competitive multi-agent game.

arXiv:2302.11457

Semantic Information Marketing in The Metaverse: A Learning-Based Contract Theory Framework

In this paper, we address the problem of designing incentive mechanisms by a virtual service provider (VSP) to hire sensing IoT devices to sell their sensing data to help creating and rendering the digital copy of the physical world in the Metaverse. Due to the limited bandwidth, we propose to use semantic extraction algorithms to reduce the delivered data by the sensing IoT devices. Nevertheless, mechanisms to hire sensing IoT devices to share their data with the VSP and then deliver the constructed digital twin to the Metaverse users are vulnerable to adverse selection problem. The adverse selection problem, which is caused by information asymmetry between the system entities, becomes harder to solve when the private information of the different entities are multi-dimensional. We propose a novel iterative contract design and use a new variant of multi-agent reinforcement learning (MARL) to solve the modelled multi-dimensional contract problem. To demonstrate the effectiveness of our algorithm, we conduct extensive simulations and measure several key performance metrics of the contract for the Metaverse. Our results show that our designed iterative contract is able to incentivize the participants to interact truthfully, which maximizes the profit of the VSP with minimal individual rationality (IR) and incentive compatibility (IC) violation rates. Furthermore, the proposed learning-based iterative contract framework has limited access to the private information of the participants, which is to the best of our knowledge, the first of its kind in addressing the problem of adverse selection in incentive mechanisms.

arXiv:2302.10825

Curiosity-driven Exploration in Sparse-reward Multi-agent Reinforcement Learning

Sparsity of rewards while applying a deep reinforcement learning method negatively affects its sample-efficiency. A viable solution to deal with the sparsity of rewards is to learn via intrinsic motivation which advocates for adding an intrinsic reward to the reward function to encourage the agent to explore the environment and expand the sample space. Though intrinsic motivation methods are widely used to improve data-efficient learning in the reinforcement learning model, they also suffer from the so-called detachment problem. In this article, we discuss the limitations of intrinsic curiosity module in sparse-reward multi-agent reinforcement learning and propose a method called I-Go-Explore that combines the intrinsic curiosity module with the Go-Explore framework to alleviate the detachment problem.

arXiv:2302.10418공식 코드 · Papers with Code

MAC-PO: Multi-Agent Experience Replay via Collective Priority Optimization

Experience replay is crucial for off-policy reinforcement learning (RL) methods. By remembering and reusing the experiences from past different policies, experience replay significantly improves the training efficiency and stability of RL algorithms. Many decision-making problems in practice naturally involve multiple agents and require multi-agent reinforcement learning (MARL) under centralized training decentralized execution paradigm. Nevertheless, existing MARL algorithms often adopt standard experience replay where the transitions are uniformly sampled regardless of their importance. Finding prioritized sampling weights that are optimized for MARL experience replay has yet to be explored. To this end, we propose MAC-PO, which formulates optimal prioritized experience replay for multi-agent problems as a regret minimization over the sampling weights of transitions. Such optimization is relaxed and solved using the Lagrangian multiplier approach to obtain the close-form optimal sampling weights. By minimizing the resulting policy regret, we can narrow the gap between the current policy and a nominal optimal policy, thus acquiring an improved prioritization scheme for multi-agent tasks. Our experimental results on Predator-Prey and StarCraft Multi-Agent Challenge environments demonstrate the effectiveness of our method, having a better ability to replay important transitions and outperforming other state-of-the-art baselines.

arXiv:2302.09605

Efficient Communication via Self-supervised Information Aggregation for Online and Offline Multi-agent Reinforcement Learning

Utilizing messages from teammates can improve coordination in cooperative Multi-agent Reinforcement Learning (MARL). Previous works typically combine raw messages of teammates with local information as inputs for policy. However, neglecting message aggregation poses significant inefficiency for policy learning. Motivated by recent advances in representation learning, we argue that efficient message aggregation is essential for good coordination in cooperative MARL. In this paper, we propose Multi-Agent communication via Self-supervised Information Aggregation (MASIA), where agents can aggregate the received messages into compact representations with high relevance to augment the local policy. Specifically, we design a permutation invariant message encoder to generate common information-aggregated representation from messages and optimize it via reconstructing and shooting future information in a self-supervised manner. Hence, each agent would utilize the most relevant parts of the aggregated representation for decision-making by a novel message extraction mechanism. Furthermore, considering the potential of offline learning for real-world applications, we build offline benchmarks for multi-agent communication, which is the first as we know. Empirical results demonstrate the superiority of our method in both online and offline settings. We also release the built offline benchmarks in this paper as a testbed for communication ability validation to facilitate further future research.

arXiv:2302.09531

AIIR-MIX: Multi-Agent Reinforcement Learning Meets Attention Individual Intrinsic Reward Mixing Network

Deducing the contribution of each agent and assigning the corresponding reward to them is a crucial problem in cooperative Multi-Agent Reinforcement Learning (MARL). Previous studies try to resolve the issue through designing an intrinsic reward function, but the intrinsic reward is simply combined with the environment reward by summation in these studies, which makes the performance of their MARL framework unsatisfactory. We propose a novel method named Attention Individual Intrinsic Reward Mixing Network (AIIR-MIX) in MARL, and the contributions of AIIR-MIX are listed as follows:(a) we construct a novel intrinsic reward network based on the attention mechanism to make teamwork more effective. (b) we propose a Mixing network that is able to combine intrinsic and extrinsic rewards non-linearly and dynamically in response to changing conditions of the environment. We compare AIIR-MIX with many State-Of-The-Art (SOTA) MARL methods on battle games in StarCraft II. And the results demonstrate that AIIR-MIX performs admirably and can defeat the current advanced methods on average test win rate. To validate the effectiveness of AIIR-MIX, we conduct additional ablation studies. The results show that AIIR-MIX can dynamically assign each agent a real-time intrinsic reward in accordance with their actual contribution.

arXiv:2302.09277

Promoting Cooperation in Multi-Agent Reinforcement Learning via Mutual Help

Multi-agent reinforcement learning (MARL) has achieved great progress in cooperative tasks in recent years. However, in the local reward scheme, where only local rewards for each agent are given without global rewards shared by all the agents, traditional MARL algorithms lack sufficient consideration of agents' mutual influence. In cooperative tasks, agents' mutual influence is especially important since agents are supposed to coordinate to achieve better performance. In this paper, we propose a novel algorithm Mutual-Help-based MARL (MH-MARL) to instruct agents to help each other in order to promote cooperation. MH-MARL utilizes an expected action module to generate expected other agents' actions for each particular agent. Then, the expected actions are delivered to other agents for selective imitation during training. Experimental results show that MH-MARL improves the performance of MARL both in success rate and cumulative reward.

arXiv:2302.10058

Differentiable Arbitrating in Zero-sum Markov Games

We initiate the study of how to perturb the reward in a zero-sum Markov game with two players to induce a desirable Nash equilibrium, namely arbitrating. Such a problem admits a bi-level optimization formulation. The lower level requires solving the Nash equilibrium under a given reward function, which makes the overall problem challenging to optimize in an end-to-end way. We propose a backpropagation scheme that differentiates through the Nash equilibrium, which provides the gradient feedback for the upper level. In particular, our method only requires a black-box solver for the (regularized) Nash equilibrium (NE). We develop the convergence analysis for the proposed framework with proper black-box NE solvers and demonstrate the empirical successes in two multi-agent reinforcement learning (MARL) environments.

arXiv:2302.08328

Learning a Multi-Agent Controller for Shared Energy Storage System

Deployment of shared energy storage systems (SESS) allows users to use the stored energy to meet their own energy demands while saving energy costs without installing private energy storage equipment. In this paper, we consider a group of building users in the community with SESS, and each user can schedule power injection from the grid as well as SESS according to their demand and real-time electricity price to minimize energy cost and meet energy demand simultaneously. SESS is encouraged to charge when the price is low, thus providing as much energy as possible for users while achieving energy savings. However, due to the complex dynamics of buildings and real-time external signals, it is a challenging task to find high-performance power dispatch decisions in real-time. By designing a multi-agent reinforcement learning framework with state-aware reward functions, SESS and users can realize power scheduling to meet the users' energy demand and SESS's charging/discharging balance without additional communication, so as to achieve energy optimization. Compared with the baseline approach without the participation of the SESS, the energy cost is saved by around 2.37% to 21.58%.

arXiv:2302.07938

Scalable Multi-Agent Reinforcement Learning with General Utilities

We study the scalable multi-agent reinforcement learning (MARL) with general utilities, defined as nonlinear functions of the team's long-term state-action occupancy measure. The objective is to find a localized policy that maximizes the average of the team's local utility functions without the full observability of each agent in the team. By exploiting the spatial correlation decay property of the network structure, we propose a scalable distributed policy gradient algorithm with shadow reward and localized policy that consists of three steps: (1) shadow reward estimation, (2) truncated shadow Q-function estimation, and (3) truncated policy gradient estimation and policy update. Our algorithm converges, with high probability, to $\epsilon$-stationarity with $\widetilde{\mathcal{O}}(\epsilon^{-2})$ samples up to some approximation error that decreases exponentially in the communication radius. This is the first result in the literature on multi-agent RL with general utilities that does not require the full observability.

arXiv:2302.07493

Adaptive incentive for cross-silo federated learning: A multi-agent reinforcement learning approach

Cross-silo federated learning (FL) is a typical FL that enables organizations(e.g., financial or medical entities) to train global models on isolated data. Reasonable incentive is key to encouraging organizations to contribute data. However, existing works on incentivizing cross-silo FL lack consideration of the environmental dynamics (e.g., precision of the trained global model and data owned by uncertain clients during the training processes). Moreover, most of them assume that organizations share private information, which is unrealistic. To overcome these limitations, we propose a novel adaptive mechanism for cross-silo FL, towards incentivizing organizations to contribute data to maximize their long-term payoffs in a real dynamic training environment. The mechanism is based on multi-agent reinforcement learning, which learns near-optimal data contribution strategy from the history of potential games without organizations' private information. Experiments demonstrate that our mechanism achieves adaptive incentive and effectively improves the long-term payoffs for organizations.