Deep reinforcement learning: a survey

Hao-nan WANG; Ning LIU; Yi-yun ZHANG; Da-wei FENG; Feng HUANG; Dong-sheng LI; Yi-ming ZHANG

doi:10.1631/FITEE.1900533

Your Location：

Home >

Browse articles >

Deep reinforcement learning: a survey

Regular Papers | Updated：2022-05-19

- Deep reinforcement learning: a survey
  Enhanced Publication
- 深度强化学习综述
- Frontiers of Information Technology & Electronic Engineering Vol. 21, Issue 12, Pages: 1726-1744(2020)
- Affiliations：
  
  Science and Technology on Parallel and Distributed Processing Laboratory, National University of Defense Technology, Changsha 410000, China
- Author bio：
  
  Hao-nan WANG, E-mail: wanghaonan14@nudt.edu.cn
  [ "Ning LIU, liuning17a@nudt.edu.cn" ]
  [ "Yi-yun ZHANG, zhangyiyun213@163.com" ]
  [ "Da-wei FENG, fengdawei@nudt.edu.cn" ]
  [ "Feng HUANG, huangfeng@nudt.edu.cn" ]
  [ "Dong-sheng LI, dsli@nudt.edu.cn" ]
  [ "Yi-ming ZHANG, zhangyiming@nudt.edu.cn" ]
- Funds：
- DOI：10.1631/FITEE.1900533
  CLC：
Scan for full text
Hao-nan WANG, Ning LIU, Yi-yun ZHANG, et al. Deep reinforcement learning: a survey. [J]. Frontiers of Information Technology & Electronic Engineering 21(12):1726-1744(2020)
DOI：

Hao-nan WANG, Ning LIU, Yi-yun ZHANG, et al. Deep reinforcement learning: a survey. [J]. Frontiers of Information Technology & Electronic Engineering 21(12):1726-1744(2020) DOI： 10.1631/FITEE.1900533.

摘要

深度强化学习已成为人工智能研究中最受欢迎的主题之一，已被广泛应用于端到端控制、机器人控制、推荐系统、自然语言对话系统等多个领域。本文对深度强化学习算法和应用进行系统分类，提供详细论述，并将现有深度强化学习算法分为基于模型的方法、无模型方法和高级深度强化学习方法。之后，全面分析探索、逆强化学习和迁移强化学习等高级算法的进展。最后，概述当前深度强化学习的代表性应用，并分析4个亟待解决的问题。

Abstract

Deep reinforcement learning (RL) has become one of the most popular topics in artificial intelligence research. It has been widely used in various fields

such as end-to-end control

robotic control

recommendation systems

and natural language dialogue systems. In this survey

we systematically categorize the deep RL algorithms and applications

and provide a detailed review over existing deep RL algorithms by dividing them into modelbased methods

model-free methods

and advanced RL methods. We thoroughly analyze the advances including exploration

inverse RL

and transfer RL. Finally

we outline the current representative applications

and analyze four open problems for future research.

关键词

强化学习深度强化学习强化学习应用

Keywords

Reinforcement learningDeep reinforcement learningReinforcement learning applications

references

P Abbeel, , , AY Ng. . Apprenticeship learning via inverse reinforcement learning. . Proc 21st Int Conf on Machine Learning, , 2004. . p.1--8. . DOI:10.1145/1015330.1015430http://doi.org/10.1145/1015330.1015430..

J Achiam, , , D Held, , , A Tamar, , , 等. . Constrained policy optimization. . Proc 34th Int Conf on Machine Learning, , 2017. . p.22--31. . ..

RRO Al-Nima, , , TT Han, , , TL Chen. . Road tracking using deep reinforcement learning for self-driving car applications. . Int Conf on Computer Recognition Systems, , 2019. . p.106--116. . DOI:10.1007/978-3-030-19738-4_12http://doi.org/10.1007/978-3-030-19738-4_12..

SO Arik, , , JT Chen, , , KN Peng, , , 等. . Neural voice cloning with a few samples. . Proc 32nd Neural Information Processing Systems, , 2018. . p.10019--10029. . ..

Y Aytar, , , T Pfaff, , , D Budden, , , 等. . Playing hard exploration games by watching YouTube. . Proc 32nd Neural Information Processing Systems, , 2018. . p.2930--2941. . ..

MG Bellemare, , , Y Naddaf, , , J Veness, , , 等. . The Arcade learning environment: an evaluation platform for general agents. . J Artif Intell Res, , 2013. . 47253--279. . DOI:10.1613/jair.3912http://doi.org/10.1613/jair.3912..

MG Bellemare, , , S Srinivasan, , , G Ostrovski, , , 等. . Unifying count-based exploration and intrinsic motivation. . Proc 30th Neural Information Processing Systems, , 2016. . p.1471--1479. . ..

MG Bellemare, , , W Dabney, , , R Munos. . A distributional perspective on reinforcement learning. . Proc 34th Int Conf on Machine Learning, , 2017. . p.449--458. . ..

C Blundell, , , J Cornebise, , , K Kavukcuoglu, , , 等. . Weight uncertainty in neural networks. . Proc 32nd Int Conf on Machine Learning, , 2015. . p.1613--1622. . ..

M Botvinick, , , S Ritter, , , JX Wang, , , 等. . Reinforcement learning, fast and slow. . Trends Cogn Sci, , 2019. . 23((5):):408--422. . DOI:10.1016/j.tics.2019.02.006http://doi.org/10.1016/j.tics.2019.02.006..

J Buckman, , , D Hafner, , , G Tucker, , , 等. . Sample-efficient reinforcement learning with stochastic ensemble value expansion. . Proc 32nd Neural Information Processing Systems, , 2018. . p.8224--8234. . ..

Y Burda, , , H Edwards, , , D Pathak, , , 等. . Large-scale study of curiosity-driven learning. . 2019. . https://arxiv.org/abs/1808.04355https://arxiv.org/abs/1808.04355, , ..

O Chapelle, , , LH Li. . An empirical evaluation of Thompson sampling. . Proc 24th Neural Information Processing Systems, , 2011. . p.2249--2257. . ..

Y Chebotar, , , K Hausman, , , M Zhang, , , 等. . Combining model-based and model-free updates for trajectory-centric reinforcement learning. . Proc 34th Int Conf on Machine Learning, , 2017. . p.703--711. . ..

L Chen, , , J Lingys, , , K Chen, , , 等. . AuTO: scaling deep reinforcement learning for datacenter-scale automatic traffic optimization. . Proc Conf of the ACM Special Interest Group on Data Communication, , 2018. . p.191--205. . DOI:10.1145/3230543.3230551http://doi.org/10.1145/3230543.3230551..

YT Chen, , , Y Assael, , , B Shillingford, , , 等. . Sample efficient adaptive text-to-speech. . 2019. . https://arxiv.org/abs/1809.10460https://arxiv.org/abs/1809.10460, , ..

K Chua, , , R Calandra, , , R McAllister, , , 等. . Deep reinforcement learning in a handful of trials using probabilistic dynamics models. . Proc 32nd Neural Information Processing Systems, , 2018. . p.4754--4765. . ..

C Devin, , , A Gupta, , , T Darrell, , , 等. . Learning modular neural network policies for multi-task and multi-robot transfer. . Proc IEEE Int Conf on Robotics and Automation, , 2017. . p.2169--2176. . DOI:10.1109/ICRA.2017.7989250http://doi.org/10.1109/ICRA.2017.7989250..

B Dhingra, , , LH Li, , , XJ Li, , , 等. . Towards end-to-end reinforcement learning of dialogue agents for information access. . Proc 55th Annual Meeting of the Association for Computational Linguistics, , 2017. . p.484--495. . DOI:10.18653/v1/P17-1045http://doi.org/10.18653/v1/P17-1045..

Y Duan, , , J Schulman, , , X Chen, , , 等. . RL2: fast reinforcement learning via slow reinforcement learning. . 2017. . https://arxiv.org/abs/1611.02779https://arxiv.org/abs/1611.02779, , ..

F Ebert, , , C Finn, , , AX Lee, , , 等. . Self-supervised visual planning with temporal skip connections. . Proc 1st Annual Conf on Robot Learning, , 2017. . p.344--356. . ..

V Feinberg, , , A Wan, , , I Stoica, , , 等. . Model-based value estimation for efficient model-free reinforcement learning. . 2018. . https://arxiv.org/abs/1803.00101https://arxiv.org/abs/1803.00101, , ..

C Finn, , , S Levine. . Deep visual foresight for planning robot motion. . Proc IEEE Int Conf on Robotics and Automation, , 2017. . p.2786--2793. . DOI:10.1109/ICRA.2017.7989324http://doi.org/10.1109/ICRA.2017.7989324..

C Finn, , , S Levine, , , P Abbeel. . Guided cost learning: deep inverse optimal control via policy optimization. . Proc 33rd Int Conf on Machine Learning, , 2016a. . p.49--58. . ..

C Finn, , , XY Tan, , , Y Duan, , , 等. . Deep spatial autoencoders for visuomotor learning. . Proc IEEE Int Conf on Robotics and Automation, , 2016b. . p.512--519. . DOI:10.1109/ICRA.2016.7487173http://doi.org/10.1109/ICRA.2016.7487173..

C Finn, , , P Abbeel, , , S Levine. . Model-agnostic meta-learning for fast adaptation of deep networks. . Proc 34th Int Conf on Machine Learning, , 2017a. . p.1126--1135. . ..

C Finn, , , T Yu, , , T Zhang, , , 等. . One-shot visual imitation learning via meta-learning. . Proc 1st Conf on Robot Learning, , 2017b. . p.357--368. . ..

M Fortunato, , , MG Azar, , , B Piot, , , 等. . Noisy networks for exploration. . 2019. . https://arxiv.org/abs/1706.10295https://arxiv.org/abs/1706.10295, , ..

J Fu, , , S Levine, , , P Abbeel. . One-shot learning of manipulation skills with online dynamics adaptation and neural network priors. . Proc IEEE/RSJ Int Conf on Intelligent Robots and Systems, , 2016. . p.4019--4026. . DOI:10.1109/IROS.2016.7759592http://doi.org/10.1109/IROS.2016.7759592..

J Fu, , , JD Co-Reyes, , , S Levine. . EX$^2$: exploration with exemplar models for deep reinforcement learning. . Proc 30$^{\rm th}$ Neural Information Processing Systems, , 2017a. . p.2577--2587. . ..

J Fu, , , K Luo, , , S Levine. . Learning robust rewards with adversarial inverse reinforcement learning. . 2017b. . https://arxiv.org/abs/1710.11248https://arxiv.org/abs/1710.11248, , ..

S Fujimoto, , , H Hoof, , , D Meger. . Addressing function approximation error in actor-critic methods. . Proc 35th Int Conf on Machine Learning, , 2018. . p.1587--1596. . ..

Y Gal, , , J Hron, , , A Kendall. . Concrete dropout. . Proc 30$^{\rm th}$ Neural Information Processing Systems, , 2017. . p.3581--3590. . ..

FM Garcia, , , PS Thomas. . A meta-MDP approach to exploration for lifelong reinforcement learning. . Proc 32$^{\rm nd}$ Neural Information Processing Systems, , 2019. . p.5691--5700. . ..

SKS Ghasemipour, , , SX Gu, , , R Zemel. . SMILe: scalable meta inverse reinforcement learning through context-conditional policies. . Proc 32nd Neural Information Processing Systems, , 2019. . p.7879--7889. . ..

JT Gu, , , H Hassan, , , J Devlin, , , 等. . Universal neural machine translation for extremely low resource languages. . Proc 16th Conf of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, , 2018a. . p.344--354. . DOI:10.18653/v1/N18-1032http://doi.org/10.18653/v1/N18-1032..

JT Gu, , , Y Wang, , , Y Chen, , , 等. . Meta-learning for low-resource neural machine translation. . Proc Conf on Empirical Methods in Natural Language Processing, , 2018b. . p.3622--3631. . DOI:10.18653/v1/D18-1398http://doi.org/10.18653/v1/D18-1398..

SX Gu, , , T Lillicrap, , , I Sutskever, , , 等. . Continuous deep Q-learning with model-based acceleration. . Proc 33rd Int Conf on Machine Learning, , 2016. . p.2829--2838. . ..

SX Gu, , , E Holly, , , T Lillicrap, , , 等. . Deep reinforcement learning for robotic manipulation with asynchronous off-policy updates. . Proc IEEE Int Conf on Robotics and Automation, , 2017a. . p.3389--3396. . DOI:10.1109/ICRA.2017.7989385http://doi.org/10.1109/ICRA.2017.7989385..

SX Gu, , , T Lillicrap, , , Z Ghahramani, , , 等. . Q-Prop: sample-efficient policy gradient with an off-policy critic. . 2017b. . https://arxiv.org/abs/1611.02247https://arxiv.org/abs/1611.02247, , ..

A Gupta, , , R Mendonca, , , YX Liu, , , 等. . Meta-reinforcement learning of structured exploration strategies. . Proc 32nd Neural Information Processing Systems, , 2018. . p.5302--5311. . ..

T Haarnoja, , , HR Tang, , , P Abbeel, , , 等. . Reinforcement learning with deep energy-based policies. . Proc 34th Int Conf on Machine Learning, , 2017. . p.1352--1361. . ..

T Haarnoja, , , A Zhou, , , P Abbeel, , , 等. . Soft actor-critic: off-policy maximum entropy deep reinforcement learning with a stochastic actor. . Proc 35th Int Conf on Machine Learning, , 2018. . p.1861--1870. . ..

M Hausknecht, , , P Stone. . Deep recurrent Q-learning for partially observable MDPs. . 2017. . https://arxiv.org/abs/1507.06527https://arxiv.org/abs/1507.06527, , ..

D He, , , YC Xia, , , T Qin, , , 等. . Dual learning for machine translation. . Proc 30th Neural Information Processing Systems, , 2016. . p.820--828. . ..

N Heess, , , S Sriram, , , J Lemmon, , , 等. . Emergence of locomotion behaviours in rich environments. . 2017. . https://arxiv.org/abs/1707.02286https://arxiv.org/abs/1707.02286, , ..

M Hessel, , , J Modayil, , , H van Hasselt, , , 等. . Rainbow: combining improvements in deep reinforcement learning. . 2018. . https://arxiv.org/abs/1710.02298https://arxiv.org/abs/1710.02298, , ..

J Ho, , , S Ermon. . Generative adversarial imitation learning. . Proc 30th Neural Information Processing Systems, , 2016. . p.4565--4573. . ..

D Horgan, , , J Quan, , , D Budden, , , 等. . Distributed prioritized experience replay. . 2018. . https://arxiv.org/abs/1803.00933https://arxiv.org/abs/1803.00933, , ..

R Houthooft, , , X Chen, , , Y Duan, , , 等. . Variational information maximizing exploration. . Proc 30$^{\rm th}$ Neural Information Processing Systems, , 2017. . p.1109--1117. . ..

S Kakade, , , J Langford. . Approximately optimal approximate reinforcement learning. . Proc 19th Int Conf on Machine Learning, , 2002. . p.267--274. . ..

D Kalashnikov, , , A Irpan, , , P Pastor, , , 等. . QT-Opt: scalable deep reinforcement learning for vision-based robotic manipulation. . Proc 2nd Conf on Robot Learning, , 2018. . p.651--673. . ..

E Klein, , , M Geist, , , B Piot, , , 等. . Inverse reinforcement learning through structured classification. . Proc 25$^{\rm th}$ Neural Information Processing Systems, , 2012. . p.1007--1015. . ..

JZ Kolter, , , AY Ng. . Near-Bayesian exploration in polynomial time. . Proc 26th Int Conf on Machine Learning, , 2009. . p.513--520. . DOI:10.1145/1553374.1553441http://doi.org/10.1145/1553374.1553441..

A Krizhevsky, , , I Sutskever, , , GE Hinton. . ImageNet classification with deep convolutional neural networks. . Proc 25th Neural Information Processing Systems, , 2012. . p.1097--1105. . ..

BJA Kröse. . Learning from delayed rewards. . Robot Auton Syst, , 1995. . 15((4):):233--235. . DOI:10.1016/0921-8890(95)00026-Chttp://doi.org/10.1016/0921-8890(95)00026-C..

S Lange, , , M Riedmiller, , , A Voigtländer. . Autonomous reinforcement learning on raw visual input data in a real world application. . Proc Int Joint Conf on Neural Networks, , 2012. . p.1--8. . DOI:10.1109/IJCNN.2012.6252823http://doi.org/10.1109/IJCNN.2012.6252823..

S Levine, , , V Koltun. . Guided policy search. . Proc 30th Int Conf on Machine Learning, , 2013. . p.1--9. . ..

S Levine, , , N Wagener, , , P Abbeel. . Learning contact-rich manipulation skills with guided policy search. . Proc IEEE Int Conf on Robotics and Automation, , 2015. . p.156--163. . DOI:10.1109/ICRA.2015.7138994http://doi.org/10.1109/ICRA.2015.7138994..

S Levine, , , C Finn, , , T Darrell, , , 等. . End-to-end training of deep visuomotor policies. . J Mach Learn Res, , 2016. . 17((1):):1334--1373. . ..

TP Lillicrap, , , JJ Hunt, , , A Pritzel, , , 等. . Continuous control with deep reinforcement learning. . Proc 4th Int Conf on Learning Representations, , 2016. . p.2829--2838. . ..

LJ Lin. . Self-improving reactive agents based on reinforcement learning, planning and teaching. . Mach Learn, , 1992. . 8((3-4):):293--321. . DOI:10.1007/BF00992699http://doi.org/10.1007/BF00992699..

HZ Mao, , , M Alizadeh, , , I Menache, , , 等. . Resource management with deep reinforcement learning. . Proc 15th ACM Workshop on Hot Topics in Networks, , 2016. . p.50--56. . DOI:10.1145/3005745.3005750http://doi.org/10.1145/3005745.3005750..

HZ Mao, , , M Schwarzkopf, , , SB Venkatakrishnan, , , 等. . Learning scheduling algorithms for data processing clusters. . Proc ACM Special Interest Group on Data Communication, , 2019a. . p.270--288. . ..

HZ Mao, , , P Negi, , , A Narayan, , , 等. . Park: an open platform for learning-augmented computer systems. . Proc 36th Int Conf on Machine Learning, , 2019b. . p.2490--2502. . ..

N Mishra, , , M Rohaninejad, , , X Chen, , , 等. . A simple neural attentive meta-learner. . 2018. . https://arxiv.org/abs/1707.03141https://arxiv.org/abs/1707.03141, , ..

V Mnih, , , K Kavukcuoglu, , , D Silver, , , 等. . Playing Atari with deep reinforcement learning. . 2013. . https://arxiv.org/abs/1312.5602https://arxiv.org/abs/1312.5602, , ..

V Mnih, , , K Kavukcuoglu, , , D Silver, , , 等. . Human-level control through deep reinforcement learning. . Nature, , 2015. . 518((7540):):529--533. . DOI:10.1038/nature14236http://doi.org/10.1038/nature14236..

V Mnih, , , AP Badia, , , M Mirza, , , 等. . Asynchronous methods for deep reinforcement learning. . Proc 33rd Int Conf on Machine Learning, , 2016. . p.1928--1937. . ..

SS Mousavi, , , M Schukat, , , E Howley. . Deep reinforcement learning: an overview. . Proc SAI Intelligent Systems Conf, , 2018. . p.426--440. . DOI:10.1007/978-3-319-56991-8_32http://doi.org/10.1007/978-3-319-56991-8_32..

O Nachum, , , M Norouzi, , , K Xu, , , 等. . Bridging the gap between value and policy based reinforcement learning. . Proc 31st Neural Information Processing Systems, , 2017a. . p.2775--2785. . ..

O Nachum, , , M Norouzi, , , K Xu, , , 等. . Trust-PCL: an off-policy trust region method for continuous control. . 2017b. . https://arxiv.org/abs/1707.01891https://arxiv.org/abs/1707.01891, , ..

A Nagabandi, , , G Kahn, , , RS Fearing, , , 等. . Neural network dynamics for model-based deep reinforcement learning with model-free fine-tuning. . IEEE Int Conf on Robotics and Automation, , 2018. . p.7559--7566. . DOI:10.1109/ICRA.2018.8463189http://doi.org/10.1109/ICRA.2018.8463189..

A Nagabandi, , , I Clavera, , , SM Liu, , , 等. . Learning to adapt in dynamic, real-world environments through meta-reinforcement learning. . 2019. . https://arxiv.org/abs/1803.11347v6https://arxiv.org/abs/1803.11347v6, , ..

AY Ng, , , SJ Russell. . Algorithms for inverse reinforcement learning. . Proc 17th Int Conf on Machine Learning, , 2000. . p.663--670. . ..

I Osband, , , C Blundell, , , A Pritzel, , , 等. . Deep exploration via bootstrapped DQN. . Proc 29$^{\rm th}$ Neural Information Processing Systems, , 2016. . p.4026--4034. . ..

G Ostrovski, , , MG Bellemare, , , A van den Oord, , , 等. . Count-based exploration with neural density models. . Proc 34th Int Conf on Machine Learning, , 2017. . p.2721--2730. . ..

E Parisotto, , , JL Ba, , , R Salakhutdinov. . Actor-Mimic: deep multitask and transfer reinforcement learning. . 2016. . https://arxiv.org/abs/1511.06342https://arxiv.org/abs/1511.06342, , ..

D Pathak, , , P Agrawal, , , AA Efros, , , 等. . Curiosity-driven exploration by self-supervised prediction. . Proc IEEE Conf on Computer Vision and Pattern Recognition Workshops, , 2017. . p.488--489. . DOI:10.1109/CVPRW.2017.70http://doi.org/10.1109/CVPRW.2017.70..

XB Peng, , , P Abbeel, , , S Levine, , , 等. . DeepMimic: example-guided deep reinforcement learning of physics-based character skills. . ACM Trans Graph, , 2018a. . 37((4):):143DOI:10.1145/3197517.3201311http://doi.org/10.1145/3197517.3201311..

XB Peng, , , M Andrychowicz, , , W Zaremba, , , 等. . Sim-to-real transfer of robotic control with dynamics randomization. . Proc IEEE Int Conf on Robotics and Automation, , 2018b. . p.3803--3810. . DOI:10.1109/ICRA.2018.8460528http://doi.org/10.1109/ICRA.2018.8460528..

W Ping, , , KN Peng, , , A Gibiansky, , , 等. . Deep voice 3: 2000-speaker neural text-to-speech. . Proc Int Conf on Learning Representations, , 2018. . p.214--217. . ..

T Pohlen, , , B Piot, , , T Hester, , , 等. . Observe and look further: achieving consistent performance on Atari. . 2018. . https://arxiv.org/abs/1805.11593https://arxiv.org/abs/1805.11593, , ..

S Racanière, , , T Weber, , , DP Reichert, , , 等. . Imagination-augmented agents for deep reinforcement learning. . Proc 31st Neural Information Processing Systems, , 2017. . p.5694--5705. . ..

R Rahmatizadeh, , , P Abolghasemi, , , A Behal, , , 等. . Learning real manipulation tasks from virtual demonstrations using LSTM. . 2016. . https://arxiv.org/abs/1603.03833v2https://arxiv.org/abs/1603.03833v2, , ..

A Rajeswaran, , , S Ghotra, , , B Ravindran, , , 等. . EPOpt: learning robust neural network policies using model ensembles. . 2017. . https://arxiv.org/abs/1610.01283https://arxiv.org/abs/1610.01283, , ..

K Rakelly, , , A Zhou, , , D Quillen, , , 等. . Efficient off-policy meta-reinforcement learning via probabilistic context variables. . Proc 36th Int Conf on Machine Learning, , 2019. . p.5331--5340. . ..

ND Ratliff, , , JA Bagnell, , , MA Zinkevich. . Maximum margin planning. . Proc 23rd Int Conf on Machine Learning, , 2006. . p.729--736. . DOI:10.1145/1143844.1143936http://doi.org/10.1145/1143844.1143936..

D Russo, , , BV Roy. . Learning to optimize via information-directed sampling. . Proc 27th Neural Information Processing Systems, , 2014. . p.1583--1591. . ..

AA Rusu, , , SG Colmenarejo, , , C Gulcehre, , , 等. . Policy distillation. . 2016a. . https://arxiv.org/abs/1511.06295https://arxiv.org/abs/1511.06295, , ..

AA Rusu, , , NC Rabinowitz, , , G Desjardins, , , 等. . Progressive neural networks. . 2016b. . https://arxiv.org/abs/1606.04671https://arxiv.org/abs/1606.04671, , ..

T Schaul, , , J Quan, , , I Antonoglou, , , 等. . Prioritized experience replay. . 2016. . https://arxiv.org/abs/1511.05952https://arxiv.org/abs/1511.05952, , ..

J Schulman, , , S Levine, , , P Moritz, , , 等. . Trust region policy optimization. . Proc Int Conf on Machine Learning, , 2015. . p.1889--1897. . ..

J Schulman, , , P Moritz, , , S Levine, , , 等. . High-dimensional continuous control using generalized advantage estimation. . 2016. . https://arxiv.org/abs/1506.02438https://arxiv.org/abs/1506.02438, , ..

J Schulman, , , F Wolski, , , P Dhariwal, , , 等. . Proximal policy optimization algorithms. . 2017. . https://arxiv.org/abs/1707.06347https://arxiv.org/abs/1707.06347, , ..

HY Shum, , , XD He, , , D Li. . From Eliza to XiaoIce: challenges and opportunities with social chatbots. . Front Inform Technol Electron Eng, , 2018. . 19((1):):10--26. . DOI:10.1631/FITEE.1700826http://doi.org/10.1631/FITEE.1700826..

D Silver, , , G Lever, , , N Heess, , , 等. . Deterministic policy gradient algorithms. . Proc 31st Int Conf on Machine Learning, , 2014. . p.387--395. . ..

D Silver, , , A Huang, , , CJ Maddison, , , 等. . Mastering the game of go with deep neural networks and tree search. . Nature, , 2016. . 529((7587):):484--489. . DOI:10.1038/nature16961http://doi.org/10.1038/nature16961..

RJ Skerry-Ryan, , , E Battenberg, , , Y Xiao, , , 等. . Towards end-to-end prosody transfer for expressive speech synthesis with Tacotron. . Int Conf on Machine Learning, , 2018. . p.4693--4702. . ..

BC Stadie, , , G Yang, , , R Houthooft, , , 等. . Some considerations on learning to explore via meta-reinforcement learning. . 2018. . https://arxiv.org/abs/1803.01118https://arxiv.org/abs/1803.01118, , ..

AL Strehl, , , ML Littman. . An analysis of model-based interval estimation for Markov decision processes. . J Comput Syst Sci, , 2008. . 74((8):):1309--1331. . DOI:10.1016/j.jcss.2007.08.009http://doi.org/10.1016/j.jcss.2007.08.009..

RS Sutton. . Learning to predict by the methods of temporal differences. . Mach Learn, , 1988. . 3((1):):9--44. . DOI:10.1023/A:1022633531479http://doi.org/10.1023/A:1022633531479..

RS Sutton, , , AG Barto. . Reinforcement Learning: an Introduction, , 2nd Ed.::Cambridge, MA, USAMIT Press, , 2018. ..

HR Tang, , , R Houthooft, , , D Foote, , , 等. . #Exploration: a study of count-based exploration for deep reinforcement learning. . Proc 31st Neural Information Processing Systems, , 2017. . p.2753--2762. . ..

H van Hasselt, , , A Guez, , , D Silver. . Deep reinforcement learning with double Q-learning. . Proc 30th AAAI Conf on Artificial Intelligence, , 2016. . p.2096--2100. . ..

J Vanschoren. . Meta-learning: a survey. . 2018. . https://arxiv.org/abs/1810.03548https://arxiv.org/abs/1810.03548, , ..

O Vinyals, , , T Ewalds, , , S Bartunov, , , 等. . StarCraft II: a new challenge for reinforcement learning. . 2017. . https://arxiv.org/abs/1708.04782https://arxiv.org/abs/1708.04782, , ..

JX Wang, , , Z Kurth-Nelson, , , D Tirumala, , , 等. . Learning to reinforcement learn. . 2017. . https://arxiv.org/abs/1611.05763https://arxiv.org/abs/1611.05763, , ..

ZY Wang, , , T Schaul, , , M Hessel, , , 等. . Dueling network architectures for deep reinforcement learning. . Proc 33rd Int Conf on Machine Learning, , 2016. . p.1995--2003. . ..

ZY Wang, , , V Bapst, , , N Heess, , , 等. . Sample efficient actor-critic with experience replay. . 2017. . https://arxiv.org/abs/1611.01224https://arxiv.org/abs/1611.01224, , ..

M Watter, , , JT Springenberg, , , J Boedecker, , , 等. . Embed to control: a locally linear latent dynamics model for control from raw images. . Proc 28th Neural Information Processing Systems, , 2015. . p.2746--2754. . ..

RJ Williams. . Simple statistical gradient-following algorithms for connectionist reinforcement learning. . Mach Learn, , 1992. . 8((3-4):):229--256. . DOI:10.1023/A:1022672621406http://doi.org/10.1023/A:1022672621406..

YH Wu, , , E Mansimov, , , RB Grosse, , , 等. . Scalable trust-region method for deep reinforcement learning using Kronecker-factored approximation. . Proc 30$^{\rm th}$ Neural Information Processing Systems, , 2017. . p.5279--5288. . ..

C Xia, , , Kamel A El. . Neural inverse reinforcement learning in autonomous navigation. . Robot Auton Syst, , 2016. . 841--14. . DOI:10.1016/j.robot.2016.06.003http://doi.org/10.1016/j.robot.2016.06.003..

A Yahya, , , A Li, , , M Kalakrishnan, , , 等. . Collective robot reinforcement learning with distributed asynchronous guided policy search. . IEEE/RSJ Int Conf on Intelligent Robots and Systems, , 2017. . p.79--86. . DOI:10.1109/IROS.2017.8202141http://doi.org/10.1109/IROS.2017.8202141..

TH Yu, , , C Finn, , , AN Xie, , , 等. . One-shot imitation from observing humans via domain-adaptive meta-learning. . 2018. . https://arxiv.org/abs/1802.01557v1https://arxiv.org/abs/1802.01557v1, , ..

WH Yu, , , J Tan, , , CK Liu, , , 等. . Preparing for the unknown: learning a universal policy with online system identification. . 2017. . https://arxiv.org/abs/1702.02453https://arxiv.org/abs/1702.02453, , ..

M Zhang, , , S Vikram, , , L Smith, , , 等. . SOLAR: deep structured representations for model-based reinforcement learning. . Proc 36th Int Conf on Machine Learning, , 2019. . p.7444--7453. . ..

BD Ziebart, , , A Maas, , , JA Bagnell, , , 等. . Maximum entropy inverse reinforcement learning. . Proc 23rd AAAI Conf on Artificial Intelligence, , 2008. . p.1433--1438. . ..

L Zintgraf, , , K Shiarli, , , V Kurin, , , 等. . Fast context adaptation via meta-learning. . Proc 36th Int Conf on Machine Learning, , 2019. . p.7693--7702. . ..

Views

Downloads

CSCD

Alert me when the article has been cited

Submit

Tools

Publicity Resources

Embedding expert demonstrations into clustering buffer for effective deep reinforcement learning

A home energy management approach using decoupling value and policy in reinforcement learning

A multipath routing algorithm for satellite networksbased on service demand and traffic awareness

Soft-HGRNs: soft hierarchical graph recurrent networks for multi-agent partially observable environments

Image-based traffic signal control via world models

Related Author

No data

Related Institution

Institute of Science and Technology for Brain-inspired Intelligence, Fudan University

Shanghai Key Laboratory of Intelligent Information Processing, School of Computer Science, Fudan University

School of Electrical and Electronics Engineering, Nanyang Technological University

School of Electrical Engineering and Telecommunications, University of New South Wales

Department of Electrical Engineering, Nantong University

Address：Zhejiang University Press, 148 Tianmushan Road, Hangzhou, China Postal code：310028
Tel：+86-571-88273162 Email：fitee@zju.edu.cn
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰