Deep Reinforcement Learning: A Brief Survey
Open accessThe DOI matches its Crossref record; 177 of the 182 references it cites were found in a registry or an index.
Abstract
Deep reinforcement learning (DRL) is poised to revolutionize the field of artificial intelligence (AI) and represents a step toward building autonomous systems with a higher-level understanding of the visual world. Currently, deep learning is enabling reinforcement learning (RL) to scale to problems that were previously intractable, such as learning to play video games directly from pixels. DRL algorithms are also applied to robotics, allowing control policies for robots to be learned directly from camera inputs in the real world. In this survey, we begin with an introduction to the general field of RL, then progress to the main streams of value-based and policy-based methods. Our survey will cover central algorithms in deep RL, including the deep Q-network (DQN), trust region policy optimization (TRPO), and asynchronous advantage actor critic. In parallel, we highlight the unique advantages of deep neural networks, focusing on visual understanding via RL. To conclude, we describe several current areas of research within the field.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Foundational surveys on RL [7,72] provide broad overviews of classical and deep RL, respectively, but do not focus on human feedback or LLMs.”
From RLHF Deciphered: A Critical Analysis of Reinforcement Learning from Human Feedback for LLMs (2026) · cites this paper as [7]
In the citing paper
“Classical approaches employ reinforcement learning (RL) [6, 7] and imitation learning [8-10] that operate on primitive actions, which could be challenging for systematic exploration [11-15], interpretability [16-18], and generalization [19-21].”
From Voyager: An Open-Ended Embodied Agent with Large Language Models (2023) · cites this paper as [7]
In the citing paper
“This can be frequently modeled as decision making and learning in a Markov decision process (MDP) [160].”
From Edge artificial intelligence for 6g: vision, enabling technologies, and applications (2022) · cites this paper as [160]
Citations
Cites
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Learning representations by back-propagating errors
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986) · Nature · doi:10.1038/323533a0 · Cited by 31,892 · cited 2×
“The workhorse of DRL, however, remains backpropagation [162, 111].”
Human-level control through deep reinforcement learning
Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., Graves, A., Riedmiller, M., Fidjeland, A. K., Ostrovski, G., et al. (2015) · Nature · doi:10.1038/nature14236 · Cited by 30,357 · cited 10×
“[34] train locally linear models for use with the NAF algorithm— the continuous equivalent of the DQN [71]—to improve the algorithm’s sample complexity in the robotic domain where”
InfluentialReinforcement Learning: An Introduction
Sutton, R. S., & Barto, A. G. (1998) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.1998.712192 · Cited by 27,225 · cited 7×
“(temporal) credit assignment problem [115].”
InfluentialReinforcement Learning: An Introduction
(2005) · IEEE Transactions on Neural Networks · doi:10.1109/tnn.2004.842673 · Cited by 24,496
Auto-Encoding Variational Bayes
Kingma, D. P., & Welling, M. (2013) · arXiv · doi:10.48550/arxiv.1312.6114 · Cited by 20,067 · cited 1×
“[22] devised a method for calculating gradients to optimize stochastic policies by “reparameterizing” [30], [60] the stochasticity away from the network, thereby allowing standard gradients to be used (instead of the high-variance REINFORCE estimator [97]).”
Labaca-Castro, R. (2023) · Machine Learning under Malware Attack · doi:10.1007/978-3-658-40442-0_9 · Cited by 19,272
GAN(Generative Adversarial Nets)
(2017) · Journal of Japan Society for Fuzzy Theory and Intelligent Informatics · doi:10.3156/jsoft.29.5_177_2 · Cited by 19,270
Mastering the game of Go with deep neural networks and tree search
Silver, D., Huang, A., Maddison, C. J., Guez, A., Sifre, L., van den Driessche, G., Schrittwieser, J., Antonoglou, I., Panneershelvam, V., Lanctot, M., et al. (2016) · Nature · doi:10.1038/nature16961 · Cited by 15,612 · cited 7×
“dynamic programming over all possible actions (Figure 3, top left), sample trajectories for heuristic search (as was done by AlphaGo [108]), or even perform an exhaustive search (Figure 3, top right).”
InfluentialDistilling the Knowledge in a Neural Network
Hinton, G., Vinyals, O., & Dean, J. (2015) · arXiv · doi:10.48550/arxiv.1503.02531 · Cited by 14,204 · cited 1×
“technique known as distillation [40] has been utilised to both transfer policies learned by large DQNs to smaller DQNs, and transfer policies learned by several DQNs trained on separate games to one single DQN [85, 95].”
Cited by
Items 701–800 of 1,021
Song, D., Gan, W., Yao, P., Zang, W., Zhang, Z., & Qu, X. (2022) · Ocean Engineering · doi:10.1016/j.oceaneng.2022.110947 · Cited by 54
Admission Control for 5G Core Network Slicing Based on Deep Reinforcement Learning
Jácome, W. F. V., Rendón, O. M. C., & da Fonseca, N. L. S. (2022) · IEEE Systems Journal · doi:10.1109/jsyst.2022.3172658 · Cited by 53
“DSARA fixes the Target NN temporarily and updates it periodically with the trained parameters of the Evaluation NN for learning stability reasons [28] [27].”
MethodAn improved marine predators algorithm for the optimal design of hybrid renewable energy systems
Houssein, E. H., Ibrahim, I. E., Kharrich, M., & Kamel, S. M. (2022) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2022.104722 · Cited by 54
Bing, Z., Zhou, H., Li, R., Su, X., Morin, F. O., Huang, K., & Knoll, A. (2022) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2022.3172754 · Cited by 58
“D EEP reinforcement learning has successfully revolutionized Rubik’s cube or enabling autonomous driving, to mind games such as AlphaGo, Atari games, and Starcraft [1]–[6].”
BackgroundDeep reinforcement learning for adaptive mesh refinement
Foucart, C., Charous, A., & Lermusiaux, P. F. J. (2023) · Journal of Computational Physics · doi:10.1016/j.jcp.2023.112381 · Cited by 71
Xiao, J. Q., Yuan, G., Xue, Y., He, J., Wang, Y., Zou, Y., & Wang, Z. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.127904 · Cited by 56
Shao, S., Tian, Y., & Zhang, Y. (2024) · Swarm and Evolutionary Computation · doi:10.1016/j.swevo.2024.101817 · Cited by 52
Multi-timescale reinforcement learning in the brain
Masset, P., Tano, P., Kim, H. R., Malik, A. N., Pouget, A., & Uchida, N. (2025) · Nature · doi:10.1038/s41586-025-08929-9 · Cited by 59
Rupprecht, T., & Wang, Y. (2022) · Neural Networks · doi:10.1016/j.neunet.2022.05.013 · Cited by 59
LIFT: Reinforcement Learning in Computer Systems by Learning From DemonstrationsPreprint
Schaarschmidt, M., Kuhnle, A., Ellis, B., Fricke, K., Gessert, F., & Yoneki, E. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.07903 · Cited by 59
Balancing a CartPole System with Reinforcement Learning -- A TutorialPreprint
Kumar, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.04938 · Cited by 52
“However, the recent success of deep learning algorithms has revived the field of reinforcement learning finding renewed interest among researchers who are now successfully applying this to solve very complex problems which were considered intractable earlier [1].”
BackgroundVisual Analytics for RNN-Based Deep Reinforcement Learning
Wang, J., Zhang, W., Yang, H., Yeh, C. M., & Wang, L. (2021) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2021.3076749 · Cited by 61
“In general, there are three RL methods to achieve this [42]: policy-based, value-based, and actor-critic RLs.”
MethodApplications of Reinforcement Learning for maintenance of engineering systems: A review
Marugán, A. P. (2023) · Advances in Engineering Software · doi:10.1016/j.advengsoft.2023.103487 · Cited by 57
Huang, X., Ng, W. L., & Yeong, W. Y. (2023) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-023-02167-4 · Cited by 52
“…basic approaches to apply machine learning for different situations; supervised learning (Caruana & Niculescu-Mizil, 2006), unsupervised learning (Ghahramani, 2004; Hastie et al., 2009), and reinforcement learning (Arulkumaran et al., 2017; Kaelbling et al., 1996; Wiering & Van Otterlo, 2012).”
Cheng, H., Li, H., Dai, Q., & Yang, J. (2023) · Chaos Solitons & Fractals · doi:10.1016/j.chaos.2023.113809 · Cited by 52
Ghorbian, M., & Ghorbian, . (2023) · Heliyon · doi:10.1016/j.heliyon.2023.e22427 · Cited by 63
“In contrast to supervised learning, reinforcement learning relies solely upon the agents ’ experiences to guide their education [72].”
BackgroundChougule, S., Chaudhari, B. S., Ghorpade, S. N., & Zennaro, M. (2024) · World Electric Vehicle Journal · doi:10.3390/wevj15020039 · Cited by 58
Heng, J., Dong, Y., Lai, L., Zhou, Z., & Frangopol, D. M. (2024) · Automation in Construction · doi:10.1016/j.autcon.2024.105697 · Cited by 53
Advancements and challenges in UAV-based communication networks: A comprehensive scholarly analysis
Wan, F., Yaseen, M. B., Riaz, M. B., Shafiq, A., Thakur, A., & Rahman, M. O. (2024) · Results in Engineering · doi:10.1016/j.rineng.2024.103271 · Cited by 54
Dai, X., Chen, R., Guan, S., Li, W.-T., & Yuen, C. (2025) · Building Simulation · doi:10.1007/s12273-025-1306-y · Cited by 56
“Therefore, RL stands out due to its strength in developing autonomous systems with a higher-level understanding of the target environment (Arulkumaran et al. 2017).”
LLM-Powered AI Agent Systems and Their Applications in Industry
Liang, G., & Tong, Q. (2025) · IEEE World AI IoT Congress (AIIoT) · doi:10.1109/aiiot65859.2025.11105299 · Cited by 70
“Following rule-based agents, RL emerged as a powerful framework for training agents to make sequential decisions by interacting with their environment [12]–[15].”
Double Q-Learning for Radiation Source Detection
Liu, Z., & Abbaszadeh, S. (2019) · Sensors · doi:10.3390/s19040960 · Cited by 51
“With the recent development of computation hardwares and deep neural networks (DNN), the combination of RL and DNN shows great success in handling complex tasks such as playing video games, driving vehicles, playing Go, and controlling robots, to name a few [22].”
MethodA Deep Reinforcement Learning-Enabled Dynamic Redeployment System for Mobile Ambulances
Ji, S., Zheng, Y., Wang, Z., & Li, T. (2019) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3314402 · Cited by 56
Survey on frontiers of language and robotics
Taniguchi, T., Mochihashi, D., Nagai, T., Uchida, S., Inoue, N., Kobayashi, I., Nakamura, T., Hagiwara, Y., Iwahashi, N., & Inamura, T. (2019) · Advanced Robotics · doi:10.1080/01691864.2019.1632223 · Cited by 60
Parvaresh, N., Kulhandjian, M., Kulhandjian, H., D’Amours, C., & Kantarcı, B. (2022) · Vehicular Communications · doi:10.1016/j.vehcom.2022.100474 · Cited by 63
Meta-Learning through Hebbian Plasticity in Random Networks
Najarro, E., & Risi, S. (2020) · IT University Of Copenhagen (IT University of Copenhagen) · doi:10.48550/arxiv.2007.02686 · Cited by 98
Smart Magnetic Microrobots Learn to Swim with Deep Reinforcement Learning
Behrens, M. R., & Ruder, W. C. (2022) · Advanced Intelligent Systems · doi:10.1002/aisy.202200023 · Cited by 52
Machine Learning for Bioelectronics on Wearable and Implantable Devices: Challenges and Potential
Goh, G. D., Lee, J. M., Goh, G. L., Huang, X., Lee, S., & Yeong, W. Y. (2022) · Tissue Engineering Part A · doi:10.1089/ten.tea.2022.0119 · Cited by 52
Gkonis, P. K. (2022) · IEEE Access · doi:10.1109/access.2022.3232855 · Cited by 57
“This is mathematically formulated as follows [43], [44]: where Q new is the updated Q value, s t is the state at the current time interval and s t + 1 is the state at the next time interval.”
BackgroundA Federated Deep Reinforcement Learning-Based Trust Model in Underwater Acoustic Sensor Networks
He, Y. Q., Han, G., Li, A., Taleb, T., Wang, C., & Yu, H. (2023) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2023.3301825 · Cited by 52
“Federated learning (FL) enables cross-regional model training while preserving the localization of training data [12].”
BackgroundAshwin, S. H., & Raj, R. N. (2023) · International Journal of Information Technology · doi:10.1007/s41870-023-01412-6 · Cited by 59
“The DRL algo-rithm requires a longer training period to learn all the driving scenarios.”
MethodInfluentialDeep reinforcement learning for financial trading using multi-modal features
Avramelou, L., Nousi, P., Passalis, N., & Tefas, A. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.121849 · Cited by 58
Shu, T., Pan, Z., Ding, Z., & Zu, Z. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124765 · Cited by 51
Li, C., Zhao, J., Pan, H., Cao, L., Guan, Q., & Xu, Z. (2025) · Engineering Structures · doi:10.1016/j.engstruct.2025.120244 · Cited by 52
Saadi, A., Abghour, N., Chiba, Z., Moussaid, K., & Ali, S. (2025) · Journal Of Big Data · doi:10.1186/s40537-025-01104-x · Cited by 59
Algorithms going wild – A review of machine learning techniques for terrestrial ecology
Cipriano, C., Noce, S., Mereu, S., & Santini, M. (2025) · Ecological Modelling · doi:10.1016/j.ecolmodel.2025.111164 · Cited by 54
Puntambekar, R., Vyas, P., Thakkar, A., & Patel, D. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.131628 · Cited by 60
Artificial intelligence and machine learning for plasmonic and surface-enhanced sensing
Geddis, A., Williams, H., Bashir, S., Malenfant, J., Dubois, C., Hamlet, L., & Masson, J. (2026) · Chemical Society Reviews · doi:10.1039/d5cs01522g · Cited by 55
Towards a Software Engineering Process for Developing Data-Driven Applications
Hesenius, M., Schwenzfeier, N., Meyer, O., Koop, W., & Gruhn, V. (2019) · International Workshop on Realizing Artificial Intelligence Synergies in Software Engineering (RAISE) · doi:10.1109/raise.2019.00014 · Cited by 59
“While for many tasks there is a wide range of techniques available to solve problems, there are different application types, which occur frequently [2], [3], [10].”
BackgroundA Dynamic Bayesian Network Approach for Analysing Topic-Sentiment Evolution
Liang, H., Ganeshbabu, U., & Thorne, T. (2020) · IEEE Access · doi:10.1109/access.2020.2979012 · Cited by 50
“Deep learning techniques [27] have recently emerged as a promising framework for sentiment classification tasks.”
BackgroundZhang, H., Wang, R., Sun, W. Y., & Zhao, H. (2021) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2021.3082986 · Cited by 53
Adaptive operator selection with reinforcement learning
Durgut, R., Aydın, M. E., & Atlı, İ. (2021) · Information Sciences · doi:10.1016/j.ins.2021.10.025 · Cited by 54
Understanding AI ecosystems in the Global South: The cases of Senegal and Cambodia
Heng, S., Tsilionis, K., Scharff, C., & Wautelet, Y. (2021) · International Journal of Information Management · doi:10.1016/j.ijinfomgt.2021.102454 · Cited by 60
Park, M.-J., Cho, H. S., Yun, T.-S., Byeon, S. J., Koo, Y. J., Yoon, S.-S., Lee, D. U., Choi, S., Park, J., Lee, J., et al. (2022) · Digest of technical papers/Digest of technical papers - IEEE International Solid-State Circuits Conference · doi:10.1109/isscc42614.2022.9731562 · Cited by 49
Distributed Reinforcement Learning for Robot Teams: a Review
Wang, Y., Damani, M., Wang, P., Cao, Y., & Sartoretti, G. (2022) · Current Robotics Reports · doi:10.1007/s43154-022-00091-8 · Cited by 64
Oyewola, D. O., Dada, E. G., Omotehinwa, T. O., Emebo, O., & Oluwagbemi, O. O. (2022) · Applied Sciences · doi:10.3390/app121910166 · Cited by 55
“It employs rewards and penalties as cues for appropriate and inappropriate conduct [61].”
BackgroundYi-kun, Y., He, J., Chen, C., & Wei, J. (2023) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2023.3274853 · Cited by 51
Houetohossou, S. C. A., Houndji, V. R., Hounmenou, C. G., Sikirou, R., & Kakaï, R. G. (2023) · Artificial Intelligence in Agriculture · doi:10.1016/j.aiia.2023.08.001 · Cited by 67
Robust Learning and Control of Time-Delay Nonlinear Systems With Deep Recurrent Koopman Operators
Han, M., Li, Z., Yin, X., & Yin, X. (2023) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2023.3328432 · Cited by 49
Wang, P., Yang, H., Han, G., Yu, R., Yang, L., Sun, G., Qi, H., Wei, X., & Zhang, Q. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3439696 · Cited by 54
“The authors of [6] also demonstrate the significant potential of deep reinforcement learning (DRL) in the field of UAV navigation.”
BackgroundA Survey of DDoS Attack and Defense Technologies in Multiaccess Edge Computing
Ma, Y., Liu, L., Liu, Z., Li, F., Xie, Q., Chen, K., Lv, C., He, Y., & Li, F. (2024) · IEEE Internet of Things Journal · doi:10.1109/jiot.2024.3490897 · Cited by 60
Cuoco, E., Cavaglià, M., Heng, I. S., Keitel, D., & Messenger, C. (2025) · Living Reviews in Relativity · doi:10.1007/s41114-024-00055-8 · Cited by 53
Deep Reinforcement Learning for Target Searching in Cognitive Electronic Warfare
You, S., Diao, M., & Gao, L. (2019) · IEEE Access · doi:10.1109/access.2019.2905649 · Cited by 51
“Historically, DRL has enabled many amazing achievements [20]–[23] in game playing, simultaneous localization and mapping (SLAM), path planning, and so forth.”
BackgroundCappart, Q., Goutierre, E., Bergman, D., & Rousseau, L.-M. (2019) · AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI)) · doi:10.1609/aaai.v33i01.33011443 · Cited by 49
Dependency-Aware Attention Control for Unconstrained Face Recognition with Image Sets
Liu, X., Kumar, B. V. K. V., Yang, C., Tang, Q., & You, J. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-01252-6_34 · Cited by 49
Virtual to Real-World Transfer Learning: A Systematic Review
Ranaweera, M., & Mahmoud, Q. H. (2021) · Electronics · doi:10.3390/electronics10121491 · Cited by 53
Koshiyama, A., Firoozye, N., & Treleaven, P. (2020) · Proceedings of the First ACM International Conference on AI in Finance · doi:10.1145/3383455.3422539 · Cited by 38
“Here we use the term governance of algorithms as a ‘catch all’ for the rules, practices and processes by which an institution directs and controls algorithms and data.”
Joint Offloading and Resource Allocation Using Deep Reinforcement Learning in Mobile Edge Computing
Zhang, X., Zhang, X., & Yang, W. (2022) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2022.3184642 · Cited by 48
“Researchers also take full advantage of DRL in handling decision-making problems with large state spaces and action spaces [23].”
BackgroundMethodGu, L., Cui, M., Xu, L., & Xu, X. (2022) · Tsinghua Science & Technology · doi:10.26599/tst.2022.9010006 · Cited by 53
AI in Human-computer Gaming: Techniques, Challenges and Opportunities
Yin, Q., Yang, J. J., Huang, K., Zhao, M., Ni, W., Liang, B., Huang, Y., Wu, S., & Wang, L. (2023) · Machine Intelligence Research · doi:10.1007/s11633-022-1384-6 · Cited by 55
“chine learning[33], which has become one of the most important decision-making techniques since the breakthrough of AlphaGo[34].”
BackgroundCherrat, E. A., Raj, S., Kerenidis, I., Shekhar, A., Wood, B., Dee, J., Chakrabarti, S., Chen, R., Herman, D., Hu, S., et al. (2023) · Quantum · doi:10.22331/q-2023-11-29-1191 · Cited by 71
“The cumulative sum of the rewards perceived during this interaction can be rewritten as , since, by definition, s T contains all previous states s t (cid:48) for all t ≤ t (cid:48) < T .”
BackgroundMany-Objective Reinforcement Learning for Online Testing of DNN-Enabled Systems
Haq, F. U., Shin, D., & Briand, L. (2023) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1109/icse48619.2023.00155 · Cited by 48
“loss, approximation-based methods [19] can be used where the expected sum of future rewards for a newly discovered state can be approximated based on known states (often with the help of state abstraction when they are too complex to directly compare).”
BackgroundRobertshaw, H. H., Karstensen, L., Jackson, B. M., Sadati, S. M. H., Rhode, K., Ourselin, S., Granados, A., & Booth, T. C. (2023) · Frontiers in Human Neuroscience · doi:10.3389/fnhum.2023.1239374 · Cited by 53
“The goal of RL is to maximise the cumulative reward over time by learning a policy that optimises the agent’s current state for a set of actions [30].”
BackgroundMoreno, J. F. C., Rizzardi, A., Sicari, S., & Coen‐Porisini, A. (2023) · Computer Networks · doi:10.1016/j.comnet.2023.110016 · Cited by 67
“Another important dimension upon which DRL frameworks can be differentiated is centered precisely in 𝜋 [7,18].”
BackgroundMethodControllable Data Generation by Deep Learning: A Review
Wang, S.-Y., Du, Y., Guo, X., Pan, B., Qin, Z. S., & Zhao, L. (2024) · ACM Computing Surveys · doi:10.1145/3648609 · Cited by 59
“Goal-oriented control optimizes the goal related to target properties and data, which is designed either based on domain knowledge [127] or estimated from observations [237].”
BackgroundBlockchain-Empowered Federated Learning: Benefits, Challenges, and Solutions
Cai, Z., Chen, J., Fan, Y., Zheng, Z., & Li, K. (2025) · IEEE Transactions on Big Data · doi:10.1109/tbdata.2025.3541560 · Cited by 61
“In[60],Luetal.usedtheDeep Q-learning(DQL)[126]methodtofacilitateclientselection fortheFLprocess.”
Multi-domain non-cooperative VNF-FG embedding: A deep reinforcement learning approach
Quang, P. T. A., Bradai, A., Singh, K., & Hadjadj‐Aoul, Y. (2019) · IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS) · doi:10.1109/infcomw.2019.8845184 · Cited by 48
Bai, Z., Shangguan, W., Cai, B., & Chai, L. (2019) · Chinese Control Conference (CCC) · doi:10.23919/chicc.2019.8866005 · Cited by 51
“eration for deep RL algorithm, an increasing number of deep RL algorithms, such as deep Q network (DQN) [2], Double DQN [11], Dueling DQN [12], etc., have been proposed and applied in various studies [13]. In addition, the algorithms are briefly explained as follows. 2.1 DQN Deep Q Network is a combination between the…”
MethodInfluentialDeep Reinforcement Learning Control for Aerobatic Maneuvering of Agile Fixed-Wing Aircraft
Clarke, S. G., & Hwang, I. (2020) · AIAA SciTech Forum · doi:10.2514/6.2020-0136 · Cited by 48
DISCO: Dynamic and Invariant Sensitive Channel Obfuscation for deep neural networks
Singh, A., Chopra, A., Garza, E., Zhang, E., Vepakomma, P., Sharma, V., & Raskar, R. (2021) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr46437.2021.01195 · Cited by 71
Distributed Learning for Wireless Communications: Methods, Applications and Challenges
Qian, L., Yang, P., Xiao, M., Dobre, O. A., Renzo, M. D., Li, J., Han, Z., Qin, Y., & Zhao, J. (2022) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2022.3156756 · Cited by 60
Artificial Intelligence (AI) and Machine Learning for Multimedia and Edge Information Processing
Seng, J. K. P., Ang, K. L.-M., Peter, E., & Mmonyi, A. C. (2022) · Electronics · doi:10.3390/electronics11142239 · Cited by 55
Autonomous vehicular overtaking maneuver: A survey and taxonomy
Lodhi, S. S., Kumar, N., & Pandey, P. K. (2023) · Vehicular Communications · doi:10.1016/j.vehcom.2023.100623 · Cited by 55
Improved Artificial Potential Field Algorithm Assisted by Multisource Data for AUV Path Planning
Xing, T., Wang, X., Ding, K., Ni, K., & Zhou, Q. (2023) · Sensors · doi:10.3390/s23156680 · Cited by 51
“Reinforcement learning, which falls under the domain of machine learning, serves as the foundation for deep reinforcement learning [42].”
BackgroundGenerative technology for human emotion recognition: A scoping review
Ma, F., Yuan, Y., Xie, Y., Ren, H., Liu, I., He, Y., Ren, F., Yu, F. R., & Ni, S. (2024) · Information Fusion · doi:10.1016/j.inffus.2024.102753 · Cited by 61
Backdoor Attack and Defense on Deep Learning: A Survey
Bai, Y., Xing, G., Wu, H., Rao, Z., Ma, C., Wang, S., Liu, X., Zhou, Y., Tang, J., Huang, K., et al. (2024) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2024.3482723 · Cited by 74
“Deep learning allows reinforcement learning to tackle decision-making problems that were previously intractable, particularly those involving high-dimensional state and action spaces [42]. a) Attack purpose: Let S † denote a set of target states and a † represent a target action.”
Review of deep learning models for crypto price prediction: Implementation and evaluation
Wu, J., Zhang, X., Huang, F., Zhou, H., & Chandra, R. (2026) · Next research. · doi:10.1016/j.nexres.2026.101513 · Cited by 52
IoT-Enabled Adaptive Context-Aware and Playful Cyber-Physical System for Everyday Energy Savings
Lu, C.-H. (2018) · IEEE Transactions on Human-Machine Systems · doi:10.1109/thms.2018.2844119 · Cited by 47
“To further improve recognition performance, combining deep learning with model-based reinforcement learning [27] has shown potential.”
BackgroundDeep Q Learning Based High Level Driving Policy Determination
Min, K., Kim, H., & Huh, K. (2018) · IEEE Intelligent Vehicles Symposium · doi:10.1109/ivs.2018.8500645 · Cited by 53
“After big success of DQN in game domain which combined with reinforcement learning and neural network, deep reinforcement learning has been studied in various ways [16].”
Backgroundde Morais, G. A. P., Marcos, L. B., Bueno, J. N. A. D., de Resende, N. F., Terra, M. H., & Grassi, V. (2020) · Control Engineering Practice · doi:10.1016/j.conengprac.2020.104630 · Cited by 48
The Societal Implications of Deep Reinforcement Learning
Whittlestone, J., Arulkumaran, K., & Crosby, M. (2021) · Journal of Artificial Intelligence Research · doi:10.1613/jair.1.12360 · Cited by 66
“…previously intractable problems by improving the performance of RL algorithms in two key ways: through representation learning , allowing them to learn directly in complicated environments, and function approximation , enabling policies with complex inputs and/or outputs (Arulkumaran et al., 2017).”
BackgroundSelf-citationReinforcement learning for pursuit and evasion of microswimmers at low Reynolds number
Borra, F., Biferale, L., Cencini, M., & Celani, A. (2022) · Physical Review Fluids · doi:10.1103/physrevfluids.7.023103 · Cited by 54
“Such neural encoding can be emulated by artificial neural networks [39].”
BackgroundEmploying reinforcement learning to enhance particle swarm optimization methods
Wu, D., & Wang, G. G. (2021) · Engineering Optimization · doi:10.1080/0305215x.2020.1867120 · Cited by 46
“Reinforcement learning can be described as a Markov decision process (MDP) (Arulkumaran et al. 2017), which consists of a set of states S , a set of actions A and an immediate/instantaneous reward function R .”
BackgroundDistributed Deep Reinforcement Learning: A Survey and a Multi-player Multi-agent Learning Toolbox
Yin, Q., Yu, T., Shen, S., Yang, J. J., Zhao, M., Ni, W., Huang, K., Liang, B., & Wang, L. (2024) · Machine Intelligence Research · doi:10.1007/s11633-023-1454-4 · Cited by 49
“Plenty of algorithms have been proposed, and readers can refer to [16], [17] for a more thorough review.”
Peng, S., Siet, S., Ugli, S. I. R., Kim, D. Y., & Park, D.-S. (2024) · Applied Sciences · doi:10.3390/app14031155 · Cited by 54
Sharma, A., & Singh, M. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109013 · Cited by 46
Intelligent Resource Adaptation for Diversified Service Requirements in Industrial IoT
Zhang, W., He, Y.-Q., Zhang, T., Ying, C., & Kang, J. (2024) · IEEE Transactions on Cognitive Communications and Networking · doi:10.1109/tccn.2024.3502493 · Cited by 47
“DRL combines reinforcement learning principles with deep neural networks to determine optimal decision strategies [30].”
Flow: Deep Reinforcement Learning for Control in SUMO
Kheterpal, N., Parvate, K., Wu, C., Kreidieh, A., Vinitsky, E., & Bayen, A. M. (2018) · EPiC series in engineering · doi:10.29007/dkzb · Cited by 59
Toward Packet Routing with Fully-distributed Multi-agent Deep Reinforcement Learning
You, X., Li, X., Xu, Y., Feng, H., & Zhao, J. (2019) · International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt) · doi:10.23919/wiopt47501.2019.9144110 · Cited by 46
Adaptive Cyber Defense Against Multi-Stage Attacks Using Learning-Based POMDP
Hu, Z., Zhu, M., & Liu, P. (2020) · ACM Transactions on Privacy and Security · doi:10.1145/3418897 · Cited by 46
Arif, M. (2021) · Journal of Information Security and Cybercrimes Research · doi:10.26735/gbtv9013 · Cited by 43
“Deep reinforcement learning [140] proposed recently will be a new area to build more autonomous systems with a high level of understanding.”
BackgroundDeep reinforcement learning towards real-world dynamic thermal management of data centers
Zhang, Q., Zeng, W., Lin, Q., Chng, C.-B., Chui, C., & Lee, P. S. (2022) · Applied Energy · doi:10.1016/j.apenergy.2022.120561 · Cited by 51
Remaining driving range prediction for electric vehicles: Key challenges and outlook
Mei, P., Karimi, H. R., Huang, C., Chen, F., & Yang, S. (2023) · IET Control Theory and Applications · doi:10.1049/cth2.12486 · Cited by 54
RAMP-Net: A Robust Adaptive MPC for Quadrotors via Physics-informed Neural Network
Sanyal, S., & Roy, K. (2023) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra48891.2023.10161410 · Cited by 57
Zhou, X., Yang, J., Li, Y.-J., Li, S., & Su, Z. (2024) · Computer Communications · doi:10.1016/j.comcom.2024.107925 · Cited by 57
Shokouhifar, M., Fanian, F., Rafsanjani, M. K., Hosseinzadeh, M., & Mirjalili, S. (2024) · Computer Science Review · doi:10.1016/j.cosrev.2024.100684 · Cited by 51
Deep reinforcement learning-based local path planning in dynamic environments for mobile robot
Tao, B., & Kim, J. (2024) · Journal of King Saud University - Computer and Information Sciences · doi:10.1016/j.jksuci.2024.102254 · Cited by 49
A Unified Smart Chinese Medicine Framework for Healthcare and Medical Services
Zhang, Q., Bai, C., Yang, L. T., Chen, Z., Li, P., & Yu, H. (2019) · IEEE Transactions on Computational Biology and Bioinformatics · doi:10.1109/tcbb.2019.2914447 · Cited by 48
“So, the deep reinforcement learning models such as the deep Q-network and the deep Q-learning are potentially used to design such an automatic question and answering system for computer-aided syndrome differentiation in smart Chinesemedicine framework [17], [18].”
MethodSLAs-Aware Online Task Scheduling Based on Deep Reinforcement Learning Method in Cloud Environment
Ran, L., Shi, X., & Shang, M. (2019) · IEEE International Conference on High Performance Computing and Communications; IEEE International Conference on Smart City; IEEE International Conference on Data Science and Systems (HPCC/SmartCity/DSS) · doi:10.1109/hpcc/smartcity/dss.2019.00209 · Cited by 51
Multi-step medical image segmentation based on reinforcement learning
Tian, Z., Si, X., Zheng, Y., Chen, Z., & Li, X. (2020) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-020-01905-3 · Cited by 52
Metadata
- Authors
- Arulkumaran, K., Deisenroth, M. P., Brundage, M., & Bharath, A. A.
- Affiliations
- University of Oxford (GB)
- Institution of Engineering and Technology (GB)
- Venue
- IEEE Signal Processing Magazine
- Published
- 2017-11
- Volume
- 34
- Issue
- 6
- Pages
- 26-38
- ISSN
- 1053-5888 (ISSN-L)
- OpenAlex
- W3100789280
- Cited by
- 4,733
- References
- 182
- Citations by year
- 2026: 444 · 2025: 855 · 2024: 806 · 2023: 749 · 2022: 590 · 2021: 511 · 2020: 388 · 2019: 260 · 2018: 122 · 2017: 6
- FWCI
- 126.58
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Signal Processing Magazine (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- UCL Discovery (University College London) (submittedVersion)
- Spiral (Imperial College London) (submittedVersion)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- Physical Sciences › Computer Science › Artificial Intelligence › Artificial Intelligence in Games
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- deep reinforcement learning, value-based methods, deep Q-network, trust region policy optimization, asynchronous advantage actor-critic, visual understanding, autonomous systems
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.