Deep Reinforcement Learning 文献综述

最新推荐文章于 2024-03-30 21:55:18 发布
最新推荐文章于 2024-03-30 21:55:18 发布
阅读量1.4k 收藏 11
点赞数 1
文章标签： Deep Reinforcement Learning  文献综述

 转载地址：http://blog.csdn.net/lqfarmer/article/details/72868471

目录

值函数

策略 

离散控制

连续控制 

多智能体

文本处理

计算机视觉

机器人

游戏

蒙特卡洛

逆强化学习

多任务和迁移学习

搜索优化

层次化学习相关的文章

值函数

  • Model-Free Episodic Control, C. Blundell et al., arXiv, 2016.
  • Safe and Efficient Off-Policy Reinforcement Learning, R. Munos et al., arXiv, 2016.
  • Deep Successor Reinforcement Learning, T. D. Kulkarni et al., arXiv, 2016.
  • Unifying Count-Based Exploration and Intrinsic Motivation, M. G. Bellemare et al., arXiv, 2016.
  • Control of Memory, Active Perception, and Action in Minecraft, J. Oh et al., ICML, 2016.
  • Dynamic Frame skip Deep Q Network, A. S. Lakshminarayanan et al., IJCAI Deep RL Workshop, 2016.
  • Hierarchical Reinforcement Learning using Spatio-Temporal Abstractions and Deep Neural Networks, R. Krishnamurthy et al., arXiv, 2016.
  • Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation, T. D. Kulkarni et al., arXiv, 2016.
  • Continuous Deep Q-Learning with Model-based Acceleration, S. Gu et al., ICML, 2016.

策略 

  • Curiosity-driven Exploration in Deep Reinforcement Learning via Bayesian Neural Networks, R. Houthooft et al., arXiv, 2016.
  • Benchmarking Deep Reinforcement Learning for Continuous Control, Y. Duan et al., ICML, 2016.
  • Learning Hand-Eye Coordination for Robotic Grasping with Deep Learning and Large-Scale Data Collection, S. Levine et al., arXiv, 2016.
  • Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization, C. Finn et al., arXiv, 2016.
  • Asynchronous Methods for Deep Reinforcement Learning, V. Mnih et al., arXiv, 2016.
  • Mastering the game of Go with deep neural networks and tree search, D. Silver et al., Nature, 2016.
  • Memory-based control with recurrent neural networks, N. Heess et al., NIPS Workshop, 2015.

离散控制

  • Model-Free Episodic Control, C. Blundell et al., arXiv, 2016.
  • Safe and Efficient Off-Policy Reinforcement Learning, R. Munos et al., arXiv, 2016.
  • Deep Successor Reinforcement Learning, T. D. Kulkarni et al., arXiv, 2016.
  • Unifying Count-Based Exploration and Intrinsic Motivation, M. G. Bellemare et al., arXiv, 2016.
  • Control of Memory, Active Perception, and Action in Minecraft, J. Oh et al., ICML, 2016.
  • Dynamic Frame skip Deep Q Network, A. S. Lakshminarayanan et al., IJCAI Deep RL Workshop, 2016.
  • Hierarchical Reinforcement Learning using Spatio-Temporal Abstractions and Deep Neural Networks, R. Krishnamurthy et al., arXiv, 2016.
  • Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation, T. D. Kulkarni et al., arXiv, 2016.
  • Deep Exploration via Bootstrapped DQN, I. Osband et al., arXiv, 2016.

连续控制 

  • Curiosity-driven Exploration in Deep Reinforcement Learning via Bayesian Neural Networks, R. Houthooft et al., arXiv, 2016.
  • Benchmarking Deep Reinforcement Learning for Continuous Control, Y. Duan et al., ICML, 2016.
  • Learning Hand-Eye Coordination for Robotic Grasping with Deep Learning and Large-Scale Data Collection, S. Levine et al., arXiv, 2016.
  • Continuous Deep Q-Learning with Model-based Acceleration, S. Gu et al., ICML, 2016.
  • Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization, C. Finn et al., arXiv, 2016.
  • Asynchronous Methods for Deep Reinforcement Learning, V. Mnih et al., arXiv, 2016.
  • Memory-based control with recurrent neural networks, N. Heess et al., NIPS Workshop, 2015.
  • Variational Information Maximisation for Intrinsically Motivated Reinforcement Learning, S. Mohamed and D. J. Rezende, arXiv, 2015.
  • Continuous control with deep reinforcement learning, T. P. Lillicrap et al., ICLR, 2016.

多智能体

  • Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks, J. N. Foerster et al., arXiv, 2016.
  • Multiagent Cooperation and Competition with Deep Reinforcement Learning, A. Tampuu et al., arXiv, 2015.

文本处理

  • Strategic Dialogue Management via Deep Reinforcement Learning, H. Cuayáhuitl et al., NIPS Workshop, 2015.
  • MazeBase: A Sandbox for Learning from Games, S. Sukhbaatar et al., arXiv, 2016.
  • Deep Reinforcement Learning with an Action Space Defined by Natural Language, J. He et al., arXiv, 2015.
  • Generating Text with Deep Reinforcement Learning, H. Guo, arXiv, 2015.
  • Language Understanding for Text-based Games Using Deep Reinforcement Learning, K. Narasimhan et al., EMNLP, 2015.
  • Listen, Attend, and Walk: Neural Mapping of Navigational Instructions to Action Sequences, H. Mei et al., arXiv, 2015.

计算机视觉

  • Model-Free Episodic Control, C. Blundell et al., arXiv, 2016.
  • Deep Successor Reinforcement Learning, T. D. Kulkarni et al., arXiv, 2016.
  • Unifying Count-Based Exploration and Intrinsic Motivation, M. G. Bellemare et al., arXiv, 2016.
  • Control of Memory, Active Perception, and Action in Minecraft, J. Oh et al., ICML, 2016.
  • Dynamic Frame skip Deep Q Network, A. S. Lakshminarayanan et al., IJCAI Deep RL Workshop, 2016.
  • Hierarchical Reinforcement Learning using Spatio-Temporal Abstractions and Deep Neural Networks, R. Krishnamurthy et al., arXiv, 2016.
  • Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation, T. D. Kulkarni et al., arXiv, 2016.

机器人

  • Curiosity-driven Exploration in Deep Reinforcement Learning via Bayesian Neural Networks, R. Houthooft et al., arXiv, 2016.
  • Benchmarking Deep Reinforcement Learning for Continuous Control, Y. Duan et al., ICML, 2016.
  • Learning Hand-Eye Coordination for Robotic Grasping with Deep Learning and Large-Scale Data Collection, S. Levine et al., arXiv, 2016.
  • Continuous Deep Q-Learning with Model-based Acceleration, S. Gu et al., ICML, 2016.
  • Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization, C. Finn et al., arXiv, 2016.
  • Asynchronous Methods for Deep Reinforcement Learning, V. Mnih et al., arXiv, 2016.
  • Memory-based control with recurrent neural networks, N. Heess et al., NIPS Workshop, 2015.
  • Towards Vision-Based Deep Reinforcement Learning for Robotic Motion Control, F. Zhang et al., arXiv, 2015.

游戏

  • Model-Free Episodic Control, C. Blundell et al., arXiv, 2016.
  • Safe and Efficient Off-Policy Reinforcement Learning, R. Munos et al., arXiv, 2016.
  • Deep Successor Reinforcement Learning, T. D. Kulkarni et al., arXiv, 2016.
  • Unifying Count-Based Exploration and Intrinsic Motivation, M. G. Bellemare et al., arXiv, 2016.
  • Control of Memory, Active Perception, and Action in Minecraft, J. Oh et al., ICML, 2016.
  • Dynamic Frame skip Deep Q Network, A. S. Lakshminarayanan et al., IJCAI Deep RL Workshop, 2016.
  • Hierarchical Reinforcement Learning using Spatio-Temporal Abstractions and Deep Neural Networks, R. Krishnamurthy et al., arXiv, 2016.
  • Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation, T. D. Kulkarni et al., arXiv, 2016.
  • Deep Exploration via Bootstrapped DQN, I. Osband et al., arXiv, 2016.
  • Learning to Communicate to Solve Riddles with Deep Distributed Recurrent Q-Networks, J. N. Foerster et al., arXiv, 2016.

蒙特卡洛

  • Mastering the game of Go with deep neural networks and tree search, D. Silver et al., Nature, 2016.
  • Better Computer Go Player with Neural Network and Long-term Prediction, Y. Tian et al., ICLR, 2016.
  • Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning, X. Guo et al., NIPS, 2014.

逆强化学习

  • Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization, C. Finn et al., arXiv, 2016.
  • Maximum Entropy Deep Inverse Reinforcement Learning, M. Wulfmeier et al., arXiv, 2015.

多任务和迁移学习

  • Actor-Mimic: Deep Multitask and Transfer Reinforcement Learning, E. Parisotto, et al., ICLR, 2016.
  • Policy Distillation, A. A. Rusu et at., ICLR, 2016.
  • ADAAPT: A Deep Architecture for Adaptive Policy Transfer from Multiple Sources, J. Rajendran et al., arXiv, 2015.
  • Universal Value Function Approximators, T. Schaul et al., ICML, 2015.

搜索优化

  • Unifying Count-Based Exploration and Intrinsic Motivation, M. G. Bellemare et al., arXiv, 2016.
  • Curiosity-driven Exploration in Deep Reinforcement Learning via Bayesian Neural Networks, R. Houthooft et al., arXiv, 2016.
  • Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation, T. D. Kulkarni et al., arXiv, 2016.
  • Deep Exploration via Bootstrapped DQN, I. Osband et al., arXiv, 2016.
  • Action-Conditional Video Prediction using Deep Networks in Atari Games, J. Oh et al., NIPS, 2015.
  • Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models, B. C. Stadie et al., arXiv, 2015.

层次化学习相关的文章

  • Deep Successor Reinforcement Learning, T. D. Kulkarni et al., arXiv, 2016.
  • Hierarchical Reinforcement Learning using Spatio-Temporal Abstractions and Deep Neural Networks, R. Krishnamurthy et al., arXiv, 2016.
  • Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation, T. D. Kulkarni et al., arXiv, 2016.

 

vivimiu
关注
Play from pixels by deep reinforcement learning
程序员光剑
09-03 2046
作者：禅与计算机程序设计艺术 1.简介 在游戏领域，经典的机器学习算法模型通常依赖于监督学习方法进行训练，这种方式需要人们预先标记好游戏中所有状态和对应的动作，然后通过这些标记数据对机器学习模型进行训练，得到可以应用到新的游戏场景中的策略。然而，在现实世界中，游戏往往缺乏足够的数据标记供机器学习模型学习
强化学习重点文献汇总
thompson的博客
03-13 5582
强化学习重点文献总结
论文笔记 Exploration for Reinforcement Learning
weixin_40247273的博客
07-30 350
摘要 本文定义并强调了在强化学习背景下的安全探索（safe exploration）问题。安全的概念和将导致毁坏因而必须避免的状态与转移有关。本文介绍了用来确定一个状态安全度的安全函数，与一个可以将系统从严峻状态变回安全状态的备份策略（backup policy）。本文还展示了一个分级探索（level-based exploration）方案，它可以生成一个综合的观察基础（comprehensive base of observation），同时支持安全约束。本文在一个燃气轮机的简化仿真上评估提出的方法。
超全！深度强化学习领域值得一读的论文列表
超级超级小天才的博客
06-15 2031
参考自：https://spinningup.openai.com/en/latest/spinningup/keypapers.html 强烈建议直接去看原文，每一篇文献都有链接 以下是深度强化学习（Deep Reinforcement Learning）中值得一读的论文，实际上，这些远远不够全面，但应该能为希望在该领域上进行研究的人提供一个有用的起点。 Model-Free RL Deep Q-Learning [1] Playing Atari with Deep Reinforcement Le.
2. 强化学习篇：Deep Successor Representation(DSR) (完善中...）
wangheng673的博客
11-28 1072
DSR学习笔记及试验 主要参考文献： [1] Gershman, Samuel J. "The successor representation: its computational logic and neural substrates."Journal of Neuroscience38.33 (2018): 7193-7200. [2]Kulkarni, Tejas D., et al. "Deep successor reinforcement learning."arX...
Deep reinforcement learning in medical imaging A literature
09-23
深度强化学习（Deep Reinforcement Learning, DRL）是人工智能领域的一个重要分支，它结合了深度学习的表征能力与强化学习的决策制定策略，使得智能系统能够在与环境的交互中自主学习最优行为。在医疗成像领域，DRL...
Deep reinforcement learning in medical imaging A literature re
09-25
随着深度学习与强化学习的结合，深度强化学习（Deep Reinforcement Learning, DRL）在解决序列决策问题时展现出了巨大的潜力。近年来，DRL已被广泛应用于医学和医疗保健领域，尤其在医疗图像分析中显示出其独特的...
2021 牛津大学：Recent Advances in Reinforcement Learning in Finance
见见大魔王
04-13 1162
Recent Advances in Reinforcement Learning in Finance1 本文概述本文是牛津大学2021年对深度强化学习在金融领域内各种决策的应用的综述文章。文章分为四个部分，第一部分是整体介绍；第二部分详细介绍了强化学习的基础知识，包括马尔科夫决策过程的定义、基于值的方法、基于策略的方法；第三部分介绍了深度强化学习的相关算法；最后一...
【综述】【合作式多智能体深度强化学习研究进展】
最新发布
wq6qeg88的博客
03-30 1156
近年来，深度强化学习在多智能体系统中取得了重大进展。在这篇综述文章中，我们重点介绍了多智能体强化学习（MARL）算法的最新方法。特别是，我们重点介绍了五种关于建模和解决合作多智能体强化学习问题的常见方法：（I）独立学习者，（II）完全可观察的批评家，（III）价值函数分解，（IV）共识，以及（IV）学习沟通。首先，我们在相关论文中详细阐述了这些方法、可能的挑战以及如何缓解这些挑战。如果适用，我们会进一步在每个类别的不同论文之间建立联系。接下来，我们将介绍MARL中一些新兴的研究领域以及相关的近期论文。
<纯干货-5>Deep Reinforcement Learning深度强化学习_论文大集合
lqfarmer的博客
06-05 1万+
本文罗列了最近放出来的关于深度强化学习（Deep Reinforcement Learning，DRL）的一些论文。文章采用人工定义的方式来进行组织，按照时间的先后进行排序，越新的论文，排在越前面。希望对大家有用，同时欢迎大家提交自己阅读过的论文。 目录 •    值函数相关的文章 •    策略相关的文章 •    离散控制相关的文章 •    连续控制相关的文章 •    文本处理
(zhuan) Deep Reinforcement Learning Papers
weixin_34129696的博客
12-04 108
  Deep Reinforcement Learning Papers   A list of recent papers regarding deep reinforcement learning. The papers are organized based on manually-defined bookmarks. They are sorted by time to see the r...
深度增强学习方向论文整理
WSN_IPv6的博客
11-28 988
一、开山鼻祖DQN Playing Atari with Deep Reinforcement Learning，V. Mnih et al., NIPS Workshop, 2013. Human-level control through deep reinforcement learning, V. Mnih et al., Nature, 2015. 二.
MADRL（多智能体深度强化学习）
热门推荐
rockray21的博客
01-23 1万+
MADRL（多智能体深度强化学习）《多Agent深度强化学习综述》 《多Agent深度强化学习综述》 从多 agent 深度强化学习中通信过程的角度对现有的多 agent 深度强化学习算法进行归纳，将其归纳为全通信集中决策、全通信自主决策、欠通信自主决策三种主流形式。 当前 DRL 的两个主要方向 – 深度 Q 网络和深度策略梯度。一种是用神经网络函数近似Q网络，另一种是将状态空间映射到动作空间的函数或者分布，策略优化的目标是寻找最优的策略映射。 两者针对的过程不同 连续动作空间的策略梯度算法分为随机策
End-to-end people detection in crowded scenes
qxin的专栏
02-17 2536
问题 operate either by scanning an image in a sliding window fashion or by classifying a discrete set of proposals. overlape 问题 解决方式是 提出了 decoding an image into a set of people detections 优点 directly ou
【深度强化学习】多智能体算法汇总
见见大魔王
04-24 1万+
本文收纳了常见的多智能体强化学习方法，并简单介绍各个算法。
强化学习中的好奇心奖励机制
小星AI
12-03 3687
参考视频：李宏毅强化学习系列 参考论文： Large-Scale Study of Curiosity-Driven Learning Curiosity-driven Exploration by Self-supervised Prediction Curiosity-driven Exploration for Mapless Navigation with Deep Reinforcement Learning Exploration by Random Network Distillation 参
MARL（multi-agent reinforcement learning）的一些边缘文章（imitation、transfer、security等）
mmc2015的专栏
11-07 5804
参考：https://github.com/LantaoYu/MARL-Papers 7.4.2、Inverse MARL [1] Cooperative inverse reinforcement learning by Hadfield-Menell D,Russell S J, Abbeel P, et al. NIPS, 2016. [2] Comparison of Mul
深度强化学习精要：Grokking Deep Reinforcement Learning
"Manning.Grokking.Deep.Reinforcement.Learning.2020.10.pdf - 一本关于深度强化学习的专业书籍，由Miguel Morales撰写，Charles Isbell, Jr.作序，由Manning出版社出版。内容涵盖深度强化学习的理论与实践，旨在...
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
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