LEARNING TO SCHEDULE COMMUNICATION IN MULTI-AGENT REINFORCEMENT LEARNING

最新推荐文章于 2023-08-04 11:50:59 发布
最新推荐文章于 2023-08-04 11:50:59 发布
阅读量4k 收藏 3
点赞数 2
分类专栏： 强化学习 深度强化学习 论文研读
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/weixin_41697507/article/details/93465924
版权
SchedNet是一种用于多智能体深度强化学习的框架，解决在有限通信带宽和共享通信介质环境下，智能体如何自我调度、编码消息和根据接收到的信息选择行动的问题。通过学习每个代理部分观察信息的重要性，SchedNet决定哪些代理有权广播信息。在合作通信和导航以及捕食者-猎物的应用中，SchedNet相对于其他机制（如无通信和简单调度）表现出显著的性能优势。
摘要由CSDN通过智能技术生成
ABSTRACT

Many real-world reinforcement learning tasks require multiple agents to make se- quential decisions under the agents’ interaction, where well-coordinated actions among the agents are crucial to achieve the target goal better at these tasks. One way to accelerate the coordination effect is to enable multiple agents to communi- cate with each other in a distributed manner and behave as a group. In this paper, we study a practical scenario when (i) the communication bandwidth is limited and (ii) the agents share the communication medium so that only a restricted num- ber of agents are able to simultaneously use the medium, as in the state-of-the-art wireless networking standards. This calls for a certain form of communication scheduling. In that re

了解本专栏 订阅专栏 解锁全文 超级会员免费看
Adam婷
关注
专栏目录
订阅专栏
多智能体强化学习：Learning to Cooperate, Compete, and Communicate
weixin_41697507的博客
09-28 1412
Multiagent environments where agents compete for resources are stepping stones on the path to AGI. Multiagent environments have two useful properties: first, there is a natural curriculum—the difficulty of the environment is determined by the skill of your
多智能体强化学习 (MultiAgent Reinforcement Learning)
最新发布
程序员光剑
06-23 1479
强化学习 (Reinforcement Learning, RL) 是一种机器学习方法，它通过智能体与环境的交互来学习最优策略。在强化学习中，智能体通过观察环境的状态，采取行动并获得奖励，从而学习如何在不同的状态下采取最优的行动。强化学习已经在许多领域得到了广泛的应用，例如游戏、机器人控制、自然语言处理等。然而，在许多实际应用中，我们需要处理多个智能体之间的交互。例如，在博弈中，每个玩家都是一个智能体，他们的行动会影响其他玩家的奖励。在机器人控制中，多个机器人需要协作完成任务。
多智能体强化学习Efficient Communication in Multi-Agent Reinforcement Learning via Variance Based Control
qq_43749262的博客
05-17 610
多代理强化学习 (MARL) 最近受到广泛关注，因为它适用于广泛的实际应用。然而，实现代理之间的有效通信一直是 MARL 中的首要问题。在这项工作中，我们提出了基于方差的控制（VBC），这是一种简单而有效的技术，可以提高 MARL 中的通信效率。通过在训练阶段限制代理之间交换的消息的方差，可以有效地消除消息中的噪声成分，同时可以保留有用的部分并供代理使用以获得更好的性能。
learning to communicate with deep multi-agent reinforcement learning
叶强的博客
09-11 928
learning to communicate with deep multi-agent reinforcement
论文笔记：Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments
a1424262219的博客
10-25 1008
Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments 2017-10-2516:38:23 　【Project Page】https://blog.openai.com/learning-to-cooperate-compete-and-communicate/ 　 ...
读书笔记15：VAIN:Attentional Multi-agent Predictive Modeling
b224618的博客
09-05 800
https://arxiv.org/pdf/1706.06122.pdf 摘要：首先介绍背景，multi-agent predictive modeling是理解很多系统的关键；然后一句话介绍现有情况，最近IN模型被提出，用来解决multi-agent physical system，然后指出这个系统的不足是计算量随着interaction的指数变化；之后提出本文提出一个随着agent数量线性变...
多智能体强化学习（MARL）近年研究概览
Paper weekly
03-15 1万+
©PaperWeekly 原创 ·作者｜李文浩学校｜华东师范大学博士生研究方向｜强化学习最近由于写论文的原因，梳理了一下近几年的多智能体强化学习（MA...
A Guide Resource for Deep Reinforcement Learning
学-> 思->用
12-24 732
A Guide Resource for Deep Reinforcement Learning 1. About this work: This deep intensive learning database was initiated by the 【 Deep Reinforcement Learning Laboratory(DeepRL-Lab) 】and was jointly c...
【三】多智能体强化学习（MARL）近年研究概览 ｛Analysis of emergent behaviors（行为分析)_、Learning communication（通信学习）｝
丨汀、的博客
07-21 2533
下面遵循综述Is multiagent deep reinforcement learning the answer or the question? A brief survey对多智能体强化学习算法的分类方法，将 MARL 算法分为以下四类： Analysis of emergent behaviors（行为分析） Learning communication（通信学习） Learning cooperation（协作学习） Agents modeling agents（智能体建模） 下面我..
顶会速递 | ICLR 2020录用论文全集
热门推荐
语言智能技术笔记簿
03-10 1万+
由深度学习三巨头Yoshua Bengio和Yann LeCun牵头创办的人工智能顶会ICLR今年最终收到2594篇投稿，共687篇论文被接收，其中48篇orals，108篇spotlights，531篇posters，录取率为26.5%，相比去年的31.4%略有下降。华人学者占比为47%。这篇博客为大家列出全部录用的论文列表，方便大家自取。
IC3Net:ICLR 2019论文代码
05-05
IC3网 该存储库包含IC3Net论文的参考实现（ICLR 2019接受），了解何时在多代理协作和竞争性任务中进行大规模沟通，可在 引用 如果您在工作中使用此代码或IC3Net，请引用以下内容： @article{singh2018learning, title={Learning when to Communicate at Scale in Multiagent Cooperative and Competitive Tasks}, author={Singh, Amanpreet and Jain, Tushar and Sukhbaatar, Sainbayar}, journal={arXiv preprint arXiv:1812.09755}, year={2018} } 独立环境版本 查找gym-starcraft ，在这个仓库： 在此存储库中找到ic3n
e325: attention怎么解决_Multi-Actor-Attention-Critic (MAAC)
weixin_32380501的博客
01-10 822
第一次写知乎文章，以前的论文整理都是自己收起来的，但这篇论文写得真的太好了，发到知乎上一方面是让自己更认真地看paper，理清楚脉络，另外也是希望如果有理解错误的地方dalao们可以指正。提前谢过。这篇文章解决的问题和18年PKU那篇 NIPS（Learning Attentional Communication for Multi-Agent Cooperation）一样，但两者的解决思路截然不...
About communication in Multi-Agent Reinforcement Learning
weixin_41697507的博客
06-22 972
Communication is one of the components of MARL and an active area of research itself, as it might influence the final performance of agents, and it affects coordination or negotiation directly. Effect...
ATOC注意交流模型
dmj3214的博客
03-16 1285
ATOC:Learning Attentional Communication for Multi-Agent Cooperation 论文链接：https://arxiv.org/pdf/1805.07733.pdf ATOC注意沟通模型，该模型学习何时需要沟通以及如何整合共享信息进行合作决策，使智能体在大规模多智能体的部分可观测分布式环境下学习高效的通信。 ATOC由一个策略网络、一个Q网络、一个注意单元和一个通信通道组成。这个模型的精彩之处在于受视觉注意力循环模型的启发，设计了一种注意力单元，它可以接
Multiple Landmark Detection using Multi-AgentReinforcement Learning基于多智能体强化学习的多重地标检测
luemeon的博客
09-27 1219
2019年，开源，少量标注
Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning-笔记
jiayoudangdang的博客
02-26 1733
Social Influence as Intrinsic Motivation for Multi-Agent Deep Reinforcement Learning 0 摘要 我们提出了一种统一的机制，通过reward对其他Agent有causal influence的Agent，在Multi-Agent强化学习（MARL）中实现协调和沟通。使用反事实推理来评估causal influence。在每个时间步长，agent都会模拟它可能采取的替代操作，并计算它们对其他agent行为的influence
多智能体强化学习经典综述A Comprehensive Survey of Multi-Agent Reinforcement Learning翻译
weixin_39989705的博客
05-14 3738
因为里面的图片实在是多，就用贴图代替了~
强化学习分享（一） DQN算法原理及实现
qq_41821562的博客
08-04 4326
（一）强化学习算法介绍DQN，顾名思义，Deep Q Learning;在传统强化学习Q-Learning的基础之上，用深度学习的神经网络来拟合函Q值函数，从而达到更好的学习效果。强化学习入门：基本思想和经典算法 - 知乎 (zhihu.com)游戏交通拥堵能源分配广告推送，机器人控制，组合与序列控制问题。目前我自己将要研究的是微电网电力资源分配问题，也是强化学习的一个小应用方向。关于。
development of multi-agent reinforcement learning
03-30
Multi-agent reinforcement learning (MARL) is a subfield of reinforcement learning (RL) that involves multiple agents learning simultaneously in a shared environment. MARL has been studied for several decades, but recent advances in deep learning and computational power have led to significant progress in the field. The development of MARL can be divided into several key stages: 1. Early approaches: In the early days, MARL algorithms were based on game theory and heuristic methods. These approaches were limited in their ability to handle complex environments or large numbers of agents. 2. Independent Learners: The Independent Learners (IL) algorithm was proposed in the 1990s, which allowed agents to learn independently while interacting with a shared environment. This approach was successful in simple environments but often led to convergence issues in more complex scenarios. 3. Decentralized Partially Observable Markov Decision Process (Dec-POMDP): The Dec-POMDP framework was introduced to address the challenges of coordinating multiple agents in a decentralized manner. This approach models the environment as a Partially Observable Markov Decision Process (POMDP), which allows agents to reason about the beliefs and actions of other agents. 4. Deep MARL: The development of deep learning techniques, such as deep neural networks, has enabled the use of MARL in more complex environments. Deep MARL algorithms, such as Deep Q-Networks (DQN) and Deep Deterministic Policy Gradient (DDPG), have achieved state-of-the-art performance in many applications. 5. Multi-Agent Actor-Critic (MAAC): MAAC is a recent algorithm that combines the advantages of policy-based and value-based methods. MAAC uses an actor-critic architecture to learn decentralized policies and value functions for each agent, while also incorporating a centralized critic to estimate the global value function. Overall, the development of MARL has been driven by the need to address the challenges of coordinating multiple agents in complex environments. While there is still much to be learned in this field, recent advancements in deep learning and reinforcement learning have opened up new possibilities for developing more effective MARL algorithms.
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包

打赏作者

Adam婷

你的鼓励将是我创作的最大动力

¥1 ¥2 ¥4 ¥6 ¥10 ¥20
扫码支付：¥1
获取中
扫码支付

您的余额不足，请更换扫码支付或充值

打赏作者

实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值