A general reinforcement learning algorithm that masters chess, shogi, and Go through self-play
Open accessThe DOI matches its Crossref record; 30 of the 30 references it cites were found in a registry or an index; 5 of the 16 sources its claims cite were checked for support.
Abstract
One program to rule them all: Computers can beat humans at increasingly complex games, including chess and Go. However, these programs are typically constructed for a particular game, exploiting its properties, such as the symmetries of the board on which it is played. Silver et al. developed a program called AlphaZero, which taught itself to play Go, chess, and shogi (a Japanese version of chess) (see the Editorial, and the Perspective by Campbell). AlphaZero managed to beat state-of-the-art programs specializing in these three games. The ability of AlphaZero to adapt to various game rules is a notable step toward achieving a general game-playing system. Science , this issue p. 1140 ; see also pp. 1087 and 1118
Overview
CoChat Summary
TL;DR
AlphaZero is a single reinforcement-learning and MCTS algorithm that, given only game rules, learned superhuman play in chess, shogi, and Go through self-play. It defeated world-champion programs Stockfish, Elmo, and AlphaGo Zero using the same algorithm and network architecture across all three games.
Key findings
- In chess, AlphaZero defeated Stockfish 8, winning 155 games and losing 6 out of 1,000; it also won matches from common human openings and 2016 TCEC opening positions.
- In shogi, AlphaZero defeated Elmo, winning 98.2% of games as black and 91.2% overall; in Go, it defeated AlphaGo Zero, winning 61% of games.
- AlphaZero first outperformed Stockfish after 4 hours (300,000 steps), Elmo after 2 hours (110,000 steps), and AlphaGo Lee after 30 hours (74,000 steps); training ran for 700,000 steps, taking about 9 hours in chess, 12 hours in shogi, and 13 days in Go.
- AlphaZero searched only about 60,000 positions per second in chess and shogi, versus 60 million for Stockfish and 25 million for Elmo, yet still defeated Stockfish with 1/10 thinking time and won 46% against Elmo with 1/100 thinking time.
Methods
- AlphaZero uses a deep neural network \(f_\theta(s)\) that outputs move probabilities \(p\) and a scalar value \(v\), trained by reinforcement learning from self-play starting from random parameters; MCTS uses these outputs to guide search, with 800 simulations per move during training.
- The same algorithm, network architecture, and hyperparameters were used for chess, shogi, and Go, with no domain knowledge except game rules; no opening book, endgame tablebases, or domain-specific heuristics were used.
- Training used 5,000 first-generation TPUs to generate self-play games and 16 second-generation TPUs to train networks; evaluation matches used 3 hours per game plus 15 seconds per move, with opponents on 44 CPU cores and AlphaZero on four first-generation TPUs and 44 CPU cores.
Limitations
- Stated: AlphaZero did not use opening books, endgame tablebases, or domain-specific heuristics; the authors note that some such techniques could further improve performance but were left for future research.
- Stated: hyperparameters were reused across games without game-specific tuning, except for exploration noise and the learning rate schedule; matches against an opening book were low in diversity, and forced randomization increased losses.
- Evident from the text: results cover only three games with separate trained instances, no cross-game transfer is demonstrated, and evaluation compares AlphaZero on TPU hardware against CPU-based opponents, so hardware and time-control differences qualify direct strength comparisons.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
3 Valid · 2 Not valid · 8 Not checked · 43 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“Apart from the representation of positions and actions described above, AlphaZero uses the same network architecture as AlphaGo Zero ( 9 ), briefly recapitulated here.”
What the sources say
- Supportedcites [9] source
The Neural Network Architecture Comparison (p. 29) identifies AlphaGo Zero’s architecture as a residual tower with policy and value heads. The citation supplies the AlphaGo Zero architecture that the citing authors say AlphaZero uses; the comparison to AlphaZero is their own statement. (LLM, conf=0.88)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“and in Go, AlphaZero first outperformed AlphaGo Lee ( 9 ) after 30 hours (74,000 steps).”
What the sources say
- Partly supportedcites [9] sourceExperimental — accuracy not yet measured
Section 2, p. 7 reports that AlphaGo Zero outperformed AlphaGo Lee, but says this happened after 36 hours, not 30. The passage does not give 74,000 steps, and it names the system AlphaGo Zero rather than AlphaZero. (LLM, conf=0.90)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The training algorithm achieved similar performance in all independent runs (see fig. S3), suggesting that the high performance of AlphaZero's training algorithm is repeatable.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“AlphaZero also defeated Stockfish when given 1 / 10 as much thinking time as its opponent (i.e. searching ~ 1 / 10 , 000 as many positions),”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“and won 46% of games against Elmo when given 1 / 100 as much time (i.e. searching ~ 1 / 40 , 000 as many positions), see Fig. 2.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
23 more claims
Create a free account to see all 56 claims and to verify the 3 references still open.
- Probably not a substantial claim. 10
- It may state two separate claims as one. 23
- The quote only partly states this claim. 1
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Games have historically served as highly effective metrics for evaluating progress in deep reinforcement learning research”
From BALROG: Benchmarking Agentic LLM and VLM Reasoning On Games (2024) · cites this paper as Silver et al., 2018
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“In 2017, AlphaGo, using the same learning algorithms in TD-Gammon and a billion times more computing power, soundly beat Ke Jie, the world Go champion, in a game that is to chess what chess is to backgammon in complexity (Silver et al., 2018).”
From Large Language Models and the Reverse Turing Test (2023) · cites this paper as Silver et al., 2018
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Most of the attention was on who won, but more important was how it won: AlphaGo was creative and had discovered new moves and board positions that were outside human experience.”
From Large Language Models and the Reverse Turing Test (2023) · cites this paper as Silver et al., 2018
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“The primary method for applying more computation at inference, common in the AlphaZero family of agents2,72, is scaling the tree search and running multiple independent search attempts.”
From Olympiad-level formal mathematical reasoning with reinforcement learning (2026) · cites this paper as [2]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Modern AI can write computer programs (Li et al., 2022) or play chess at super-human level (Silver et al., 2018), but the ability of AI to perceive and act in the world remains far below human level.”
From Scaling Instructable Agents Across Many Simulated Worlds (2024) · cites this paper as Silver et al., 2018
1 more claim about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
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
Self-citationAdvances in neural information processing systems 7
(1996) · Computers & Mathematics with Applications · doi:10.1016/s0898-1221(96)90463-0 · Cited by 14,281
Identity mappings in deep residual networks
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46493-0_38 · Cited by 10,741
Identity Mappings in Deep Residual Networks
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · arXiv · doi:10.48550/arxiv.1603.05027 · Cited by 10,665
Mastering the game of Go without human knowledge
Silver, D., Schrittwieser, J., Simonyan, K., Antonoglou, I., Huang, A., Guez, A., Hubert, T., Baker, L., Lai, M., Bolton, A., et al. (2017) · Nature · doi:10.1038/nature24270 · Cited by 9,084
Self-citationIn-Datacenter Performance Analysis of a Tensor Processing Unit
Jouppi, N. P., Young, C., Patil, N., Patterson, D., Agrawal, G., Bajwa, R., Bates, S., Bhatia, S., Boden, N., Borchers, A., et al. (2017) · Proceedings of the 44th Annual International Symposium on Computer Architecture · doi:10.1145/3079856.3080246 · Cited by 4,537
Encyclopedia of Artificial Intelligence
(2009) · IGI Global eBooks · doi:10.4018/978-1-59904-849-9 · Cited by 1,396
Campbell, M., Hoane, A. J., & Hsu, F.-H. (2002) · Artificial Intelligence · doi:10.1016/s0004-3702(01)00129-1 · Cited by 1,226
Lim, T.-S., Loh, W.-Y., & Shih, Y.-S. (2000) · Machine Learning · doi:10.1023/a:1007608224229 · Cited by 1,133
An analysis of alpha-beta pruning
Knuth, D. E., & Moore, R. W. (1975) · Artificial Intelligence · doi:10.1016/0004-3702(75)90019-3 · Cited by 1,004
Research works printed but not linked: 7
- C. J. Maddison A. Huang I. Sutskever D. Silver paper presented at the International Conference on Learning Representations 2015 San Diego CA 7 to 9 May 2015.
- D. N. L. Levy M. Newborn How Computers Play Chess (Ishi Press 2009).
- W. Steinitz The Modern Chess Instructor (Edition Olms 1990).
- O. Arenz “Monte Carlo chess ” master’s thesis Technische Universität Darmstadt (2012).
- 10.1023/a:1007634325138 · DOI
- T. Kaneko K. Hoki in Advances in Computer Games – 13th International Conference ACG 2011 Revised Selected Papers Tilburg Netherlands 20 to 22 November 2011 (Springer 2012) pp. 158–169.
- R. Ramanujan A. Sabharwal B. Selman in Proceedings of the 26th Conference on Uncertainty in Artificial Intelligence (UAI 2010) Catalina Island CA 8 to 11 July (AUAI Press 2010).
Cited by
Items 601–700 of 3,632
Optimal active particle navigation meets machine learning (a)
Nasiri, M., Löwen, H., & Liebchen, B. (2023) · Europhysics Letters (EPL) · doi:10.1209/0295-5075/acc270 · Cited by 42
Finding the ground state of spin Hamiltonians with reinforcement learning
Mills, K., Ronagh, P., & Tamblyn, I. (2020) · Nature Machine Intelligence · doi:10.1038/s42256-020-0226-x · Cited by 45
Emergent Linear Representations in World Models of Self-Supervised Sequence Models
Nanda, N., Lee, A., & Wattenberg, M. (2023) · Proceedings of the 6th BlackboxNLP Workshop: Analyzing and Interpreting Neural Networks for NLP · doi:10.18653/v1/2023.blackboxnlp-1.2 · Cited by 55
Thorstad, D. (2024) · doi:10.1093/oso/9780198886143.001.0001 · Cited by 40
Artificial Intelligence, Sensors and Vital Health Signs: A Review
Junaid, S. B., Imam, A. A., Shuaibu, A. N., Basri, S. B., Kumar, G., Surakat, Y. A., Balogun, A. O., Abdulkarim, M., Garba, A., Sahalu, Y., et al. (2022) · Applied Sciences · doi:10.3390/app122211475 · Cited by 51
Chen, L., Meng, F., & Zhang, Y. (2023) · IEEE Transactions on Sustainable Computing · doi:10.1109/tsusc.2023.3251302 · Cited by 32
“chess, shogi, Go [13], [14], and simulated robots [15], [16], [17].”
Augmented Education in the Global Age: Artificial Intelligence and the Future of Learning and Work
Araya, D., & Marber, P. (2023) · Routledge eBooks · doi:10.4324/9781003230762 · Cited by 30
Kong, X., Huang, Y. C., Zhu, J., Man, X., Liu, Y., Feng, C., Gou, P., Tang, M., Wei, S., & Liu, L. (2023) · Proceedings - International Symposium on Computer Architecture/Proceedings · doi:10.1145/3579371.3589081 · Cited by 32
“RL and MCTS have yielded brilliant results in chess and computer games in recent years[57, 59].”
BackgroundMethodIs artificial consciousness achievable? Lessons from the human brain
Farisco, M., Evers, K., & Changeux, J. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106714 · Cited by 61
Towards a new paradigm in intelligence-driven computational fluid dynamics simulations
Chen, X., Wang, Z., Deng, L., Yan, J., Gong, C., Yang, B., Wang, Q., Zhang, Q., Yang, L., Pang, Y., et al. (2024) · Engineering Applications of Computational Fluid Mechanics · doi:10.1080/19942060.2024.2407005 · Cited by 36
“When considering intelligent decision-making, deep reinforcement learning models like AlphaGo and AlphaZero (Silver et al., 2018) have surpassed human capabilities in traditional board games, significantly advancing our understanding of how AI agents can acquire the ability to make complex decisions in a diverse range…”
Deep Controlled Learning for Inventory Control
Temizöz, T., Imdahl, C., Dijkman, R., Lamghari-Idrissi, D., & van Jaarsveld, W. L. (2025) · European Journal of Operational Research · doi:10.1016/j.ejor.2025.01.026 · Cited by 55
“In this vein, AlphaZero provides a relevant comparative framework.”
BackgroundMethodInfluentialAdvancements in ultrafast photonics: confluence of nonlinear optics and intelligent strategies
Wu, Q. Y., Peng, L., Huang, Z., Liu, X., Luo, M., Gao, D., & Meng, H. (2025) · Light Science & Applications · doi:10.1038/s41377-024-01732-7 · Cited by 37
“As a rapidly growing branch of machine learning, DRL is a goal-oriented algorithm, learning from interactions with the environment and guided by objectives 228,229 .”
Deep learning, transparency, and trust in human robot teamwork
Lewis, M., Li, H., & Sycara, K. P. (2020) · Elsevier eBooks · doi:10.1016/b978-0-12-819472-0.00014-9 · Cited by 35
Artificial Intelligence and the Future of Work
Benhamou, S. (2020) · Revue d économie industrielle · doi:10.4000/rei.8727 · Cited by 69
Reinforcement-learning-based matter-wave interferometer in a shaken optical lattice
Chih, L.-Y., & Holland, M. (2021) · Physical Review Research · doi:10.1103/physrevresearch.3.033279 · Cited by 32
Simultaneously constraining cosmology and baryonic physics via deep learning from weak lensing
Lu, T., Haiman, Z., & Matilla, J. M. Z. (2022) · Monthly Notices of the Royal Astronomical Society · doi:10.1093/mnras/stac161 · Cited by 29
Artificial Intelligence Methodologies for Data Management
Serey, J., Quezada, L., Alfaro, M. D., Fuertes, G., Vargas, M., Ternero, R., Sabattin, J., Durán, C., & Gutiérrez, S. (2021) · Symmetry · doi:10.3390/sym13112040 · Cited by 36
Multilevel development of cognitive abilities in an artificial neural network
Volzhenin, K., Changeux, J., & Dumas, G. (2022) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2201304119 · Cited by 42
“While the tasks considered here may appear minimalistic compared to recent sensational breakthroughs coming from artificial intelligence (53, 54) our goal was not to achieve behavioral complexity but to uncover general biological principles through natural plausibility.”
BackgroundWang, X., & Fang, X. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.120643 · Cited by 32
Planning on the Verge of AI, or AI on the Verge of Planning
Sanchez, T. W. (2023) · Urban Science · doi:10.3390/urbansci7030070 · Cited by 41
“Instead, they use Monte Carlo simulation and other methods to approximate probabilities of actions [44].”
MethodINT: An Inequality Benchmark for Evaluating Generalization in Theorem Proving
Wu, Y., Jiang, A. Q., Ba, J., & Grosse, R. (2020) · arXiv · doi:10.48550/arxiv.2007.02924 · Cited by 30
“Following AlphaGo Zero/AlphaZero [27, 28], we trained a value network for estimating the value of a state, instead of sampling rollouts to the end of the episode.”
MethodA multi-agent simulator for generating novelty in monopoly
Kejriwal, M., & Thomas, S. (2021) · Simulation Modelling Practice and Theory · doi:10.1016/j.simpat.2021.102364 · Cited by 38
Reimagining chess with AlphaZero
Tomašev, N., Paquet, U., Hassabis, D., & Kramnik, V. (2022) · Communications of the ACM · doi:10.1145/3460349 · Cited by 30
Self-citationLiu, J., Chen, R., Lou, J., Wu, H., You, Y., & Chen, Z. (2023) · Aerospace Science and Technology · doi:10.1016/j.ast.2023.108737 · Cited by 31
Chen, L., Lei, Y., Jin, S., Zhang, Y., & Zhang, L. (2024) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2024.3400189 · Cited by 52
“The rapid development of large language models (LLMs) has enabled us to obtain human-level responses across a broad range of professional fields, including history, math, engineering, etc., by providing appropriate prompts [7], [8].”
DSQN: Robust path planning of mobile robot based on deep spiking Q-network
Kumar, A., Zhang, L., Bilal, H., Wang, S., Shaikh, A. M., Lü, B., Rohra, A., & Khalid, A. (2025) · Neurocomputing · doi:10.1016/j.neucom.2025.129916 · Cited by 43
Ionotronic Neuromorphic Devices for Bionic Neural Network Applications
Yu, F., & Zhu, L. Q. (2019) · physica status solidi (RRL) - Rapid Research Letters · doi:10.1002/pssr.201800674 · Cited by 30
Computational design of syntheses leading to compound libraries or isotopically labelled targets
Molga, K., Dittwald, P., & Grzybowski, B. A. (2019) · Chemical Science · doi:10.1039/c9sc02678a · Cited by 30
Photonic architecture for reinforcement learning
Flamini, F., Hamann, A., Jerbi, S., Trenkwalder, L. M., Nautrup, H. P., & Briegel, H. J. (2020) · New Journal of Physics · doi:10.1088/1367-2630/ab783c · Cited by 37
“On the brink of this new technological revolution [2], reinforcement learning (RL) has emerged as a powerful and flexible tool to enable problem solving at an unprecedented scale [3–7].”
BackgroundHainc, N., Mannil, M., Anagnostakou, V., Alkadhi, H., Blüthgen, C., Wacht, L., Bink, A., Husain, S., Kulcsár, Z., & Winklhofer, S. F.-X. (2020) · The Neuroradiology Journal · doi:10.1177/1971400920937647 · Cited by 30
“Deep learning has recently achieved performance improvements in diverse fields such as image classification, speech recognition, natural language processing and playing games.(33) Deep learning systems encode features by using an architecture of artificial neural networks, an approach consisting of connected nodes ins…”
BackgroundA Promising Path Towards Autoformalization and General Artificial Intelligence
Szegedy, C. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-53518-6_1 · Cited by 29
Tang, Z., Zhu, Y., Zhao, D., & Lucas, S. M. (2020) · IEEE Transactions on Games · doi:10.1109/tg.2020.3022698 · Cited by 31
“[3] D. Silver et al., “A general reinforcement learning algorithm that masters Chess, Shogi, and Go through self-play,” Science, vol. 362, no. 6419, pp. 1140–1144, 2018.”
MethodMonte Carlo Tree Search as an intelligent search tool in structural design problems
Rossi, L., Winands, M. H. M., & Butenweg, C. (2021) · Engineering With Computers · doi:10.1007/s00366-021-01338-2 · Cited by 30
Monte Carlo Scene Search for 3D Scene Understanding
Hampali, S., Stekovic, S., Sarkar, S. D., Kumar, C. S., Fraundorfer, F., & Lepetit, V. (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/cvpr46437.2021.01359 · Cited by 37
Zhang, H., & Yu, T. (2020) · Springer eBooks · doi:10.1007/978-981-15-4095-0_15 · Cited by 31
Data-efficient deep reinforcement learning with expert demonstration for active flow control
Zheng, C., Xie, F., Ji, T., Zhang, X., Lu, Y., Zhou, H., & Zheng, Y. (2022) · Physics of Fluids · doi:10.1063/5.0120285 · Cited by 29
Towards Learning to Play Piano with Dexterous Hands and Touch
Xu, H., Luo, Y., Wang, S., Darrell, T., & Calandra, R. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9981221 · Cited by 50
“Reinforcement Learning (RL) has been proved to outperform humans in games such as Atari [1], Go [2] and achieve impressive results on continuous control tasks [3].”
BackgroundHuman-in-the-Loop Deep Reinforcement Learning with Application to Autonomous DrivingPreprint
Wu, J., Huang, Z., Huang, C., Hu, Z., Hang, P., Xing, Y., & Lv, C. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2104.07246 · Cited by 66
“DRL, which is another data-driven self-optimization-based algorithm, shows great potential to mitigate the aforementioned issues (11-13).”
BackgroundArtificial intelligence as structural estimation: Deep Blue, Bonanza, and AlphaGo
Igami, M. (2020) · The Econometrics Journal · doi:10.1093/ectj/utaa005 · Cited by 28
Damassino, N. (2020) · Minds and Machines · doi:10.1007/s11023-020-09551-6 · Cited by 32
A Survey on Noncooperative Games and Distributed Nash Equilibrium Seeking over Multi-Agent Networks
Yi, P., Lei, J., Li, X., Liang, S., Meng, M., & Chen, J. (2022) · CAAI Artificial Intelligence Research · doi:10.26599/air.2022.9150002 · Cited by 30
Pham, M. D., D’Angiulli, A., Dehnavi, M. M., & Chhabra, R. (2023) · Brain Sciences · doi:10.3390/brainsci13091316 · Cited by 33
“It might be refreshing to use an example that is already old news in the field of AI: the complete domination of AlphaGo software [158–160] in board games, already exhibiting behaviors that are perceived by champion gamers as “incomprehensible” and “alien” [161].”
MethodMusaddiq, A., Olsson, T., & Ahlgren, F. (2023) · Sensors · doi:10.3390/s23198263 · Cited by 38
Assessing the value of deep reinforcement learning for irrigation scheduling
Kelly, T. D., Foster, T., & Schultz, D. M. (2024) · Smart Agricultural Technology · doi:10.1016/j.atech.2024.100403 · Cited by 53
Wu, D., Zhang, R., Pore, A., Dall’Alba, D., Ha, X. T., Li, Z., Zhang, Y., Gonzalez-Herrera, F., Ourak, M., Kowalczyk, W., et al. (2024) · Biomedical Signal Processing and Control · doi:10.1016/j.bspc.2024.106179 · Cited by 44
Real-time local path planning strategy based on deep distributional reinforcement learning
Du, S., Zhu, Z., Wang, X., Han, H., & Qiao, J. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.128085 · Cited by 33
Wei, L., Li, Q., Song, Y., Stefanov, S., Dong, R., Fu, N., Siriwardane, E. M. D., Chen, F., & Hu, J. (2024) · Advanced Science · doi:10.1002/advs.202304305 · Cited by 49
Moneta, N., Grossman, S., & Schuck, N. W. (2024) · Trends in Neurosciences · doi:10.1016/j.tins.2024.10.005 · Cited by 45
Reinforcement learning-based model predictive control for greenhouse climate control
Mallick, S., Airaldi, F., Dabiri, A., Sun, C., & Schutter, B. D. (2024) · Smart Agricultural Technology · doi:10.1016/j.atech.2024.100751 · Cited by 44
Data-driven wind farm flow control and challenges towards field implementation: A review
Göçmen, T., Liew, J., Kadoche, E., et al. (2025) · Renewable and Sustainable Energy Reviews · doi:10.1016/j.rser.2025.115605 · Cited by 28
Comparison of Deep Neural Networks and Deep Hierarchical Models for Spatio-Temporal Data
Wikle, C. K. (2019) · Journal of Agricultural Biological and Environmental Statistics · doi:10.1007/s13253-019-00361-7 · Cited by 28
Five Challenges in Cloud-enabled Intelligence and Control
Abdelzaher, T., Hao, Y., Jayarajah, K., Misra, A., Skarin, P., Yao, S., Weerakoon, D., & Årzén, K.-E. (2020) · ACM Transactions on Internet Technology · doi:10.1145/3366021 · Cited by 36
“The major successes of RL have, however, been found for applications where the state space and the action space are discrete, e.g., different game playing applications such as Atari [15], AlphaGo for playing Go [19] and AlphaZero for playing chess [18].”
BackgroundZhang, Y., Chu, B., & Shu, Z. (2019) · IFAC-PapersOnLine · doi:10.1016/j.ifacol.2019.12.669 · Cited by 28
Explore and Exploit with Heterotic Line Bundle Models
Larfors, M., & Schneider, R. (2020) · Fortschritte der Physik · doi:10.1002/prop.202000034 · Cited by 50
“Notably are the dominant wins against the world champion in GO, a board game with (10170) possible board position, achieved by AlphaZero.”
BackgroundInfluentialAutomated Flowsheet Synthesis Using Hierarchical Reinforcement Learning: Proof of Concept
Göttl, Q., Tönges, Y., Grimm, D. G., & Burger, J. (2021) · Chemie Ingenieur Technik · doi:10.1002/cite.202100086 · Cited by 33
Zeng, K., Linot, A. J., & Graham, M. D. (2022) · Proceedings of the Royal Society A Mathematical Physical and Engineering Sciences · doi:10.1098/rspa.2022.0297 · Cited by 44
“Many of its milestone achievements involve defeating the best professional players in complex games such as GO [1], DOTA II [2], and Starcraft II [3], as well as the leading GO, Chess, and Shogi engines [4].”
BackgroundHybrid Reinforcement Learning for Optimal Control of Non-Linear Switching System
Li, X., Dong, L., Xue, L., & Sun, C. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3156287 · Cited by 31
“Restrictions apply. decision-making problems [20]–[22].”
On the Use of Quantum Reinforcement Learning in Energy-Efficiency Scenarios
Andrés, E., Cuéllar, M. P., & Navarro, G. (2022) · Energies · doi:10.3390/en15166034 · Cited by 36
Zhang, Y., Negenborn, R. R., & Atasoy, B. (2023) · Transportation Research Part C Emerging Technologies · doi:10.1016/j.trc.2023.104355 · Cited by 38
Xi, M., Yang, J., Wen, J., Li, Z., Lu, W. B., & Gao, X. (2023) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2023.3332172 · Cited by 34
Progress and summary of reinforcement learning on energy management of MPS-EV
Lin, Y., Chu, L., Hu, J., Hou, Z., Li, J., Jiang, J., & Zhang, Y. (2023) · Heliyon · doi:10.1016/j.heliyon.2023.e23014 · Cited by 31
“In recent years, reinforcement learning has achieved beyond human-level results on a range of domains such as Go [22] and Google Research Football [23], which demonstrated its potential for application to complex control problems.”
BackgroundCroll, H. C., Ikuma, K., Ong, S. K., & Sarkar, S. (2024) · Water Research · doi:10.1016/j.watres.2024.122179 · Cited by 38
A Reinforcement Learning Model Based on Temporal Difference Algorithm
Li, X., Lv, Z., Wang, S., Wei, Z., & Wu, L. (2019) · IEEE Access · doi:10.1109/access.2019.2938240 · Cited by 34
“The notable studied games include Go, Chess, Shougi, Multiplayer poker, and so on [4]–[10].”
BackgroundMethodAdversarial Skill Networks: Unsupervised Robot Skill Learning from Video
Mees, O., Merklinger, M., Kalweit, G., & Burgard, W. (2020) · 2020 IEEE International Conference on Robotics and Automation (ICRA) · doi:10.1109/icra40945.2020.9196582 · Cited by 26
Zhang, J., Zhang, S., Chen, L., & Liu, X. (2020) · Physical review. E · doi:10.1103/physreve.101.042402 · Cited by 29
Yager, K. G., Majewski, P. W., Noack, M. M., & Fukuto, M. (2023) · Nanotechnology · doi:10.1088/1361-6528/acd25a · Cited by 29
“This reinforcement learning approach has been highly successful in game-playing [ 102, 103 ] , for instance, and shows promise for rapid decision-making in experimental contexts [ 16, 41 ] .”
A strategic decision-making architecture toward hybrid teams for dynamic competitive problems
Bayrak, A. E., McComb, C., Cagan, J., & Kotovsky, K. (2021) · Decision Support Systems · doi:10.1016/j.dss.2020.113490 · Cited by 33
Dehghani, M., Kolahdouz-Rahimi, S., Tisi, M., & Tamzalit, D. (2022) · Software & Systems Modeling · doi:10.1007/s10270-022-00977-3 · Cited by 27
“RL does not require a large training set and has been very successful in areas where specific goals need to be achieved in a given environment [27].”
BackgroundA Systematic Review of Wi-Fi and Machine Learning Integration with Topic Modeling Techniques
Atzeni, D., Bacciu, D., Mazzei, D., & Prencipe, G. (2022) · Sensors · doi:10.3390/s22134925 · Cited by 36
The neural architecture of theory-based reinforcement learning
Tomov, M. S., Tsividis, P. A., Pouncy, T., Tenenbaum, J. B., & Gershman, S. J. (2023) · Neuron · doi:10.1016/j.neuron.2023.01.023 · Cited by 34
Bias-reduced hindsight experience replay with virtual goal prioritization
Manela, B., & Biess, A. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.02.090 · Cited by 28
“Deep reinforcement learning, the combination of reinforcement learning [1] with deep learning [2] has led to many breakthroughs in recent years in generating goal-directed behavior in artificial agents ranging from playing Atari games without prior knowledge and human guidance [3], to teaching an animated humanoid age…”
BackgroundMulti-Dimensional Urban Sensing in Sparse Mobile Crowdsensing
Liu, W., Yang, Y., Wang, E., Wang, L., Zeghlache, D., & Zhang, D. (2019) · IEEE Access · doi:10.1109/access.2019.2924184 · Cited by 26
“[22], [23] applied RL and presented the famous AlphaGo, which was the first program to defeat world-class players in Go.”
MethodReinforcement Learning for Blood Glucose Control: Challenges and Opportunities
Fox, I., & Wiens, J. (2019) · Cited by 48
Zhang, M., Tian, G., Zhang, Y., & Duan, P. (2020) · Knowledge-Based Systems · doi:10.1016/j.knosys.2020.106605 · Cited by 28
Combining Q-Learning and Search with Amortized Value EstimatesPreprint
Hamrick, J. B., Bapst, V., Sánchez‐González, Á., Pfaff, T., Weber, T., Buesing, L., & Battaglia, P. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.02807 · Cited by 45
“…heart of reinforcement learning (RL) since its inception (Bellman, 1957), and have recently seen a resurgence in the era of deep learning, with powerful function approximators inspiring a variety of effective new approaches (Silver et al., 2018; Chua et al., 2018; Hamrick, 2019; Wang et al., 2019).”
BackgroundMethodInfluentialLiu, Y., Shi, S., Wang, J., & Zhong, Y. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01547 · Cited by 48
“RL techniques have achieved success in various areas, including board games, video games, and mathematics [27, 35, 14].”
BackgroundReinforcement learning tutor better supported lower performers in a math task
Ruan, S., Nie, A., Steenbergen, W., He, J., Zhang, J., Guo, M., Liu, Y., Nguyen, K. D., Wang, C. Y., Ying, R., et al. (2024) · Machine Learning · doi:10.1007/s10994-023-06423-9 · Cited by 44
“Reinforcement learning has seen impressive successes in areas like robotics (Levine et al., 2016) and game playing (Silver et al., 2018).”
Machine learning enhanced evaluation of semiconductor quantum dots
Corcione, E., Jakob, F., Wagner, L., Joos, R., Bisquerra, A., Schmidt, M., Wieck, A. D., Ludwig, A., Jetter, M., Portalupi, S. L., et al. (2024) · Scientific Reports · doi:10.1038/s41598-024-54615-7 · Cited by 34
“7, this implies, the information contained in the data vector u is first encoded into a lower-dimensional latent feature vector ξ ∈ � ⊂ U and subsequently decoded again, in order to produce a reconstruction ˆ u ∈ U of the original input.”
BackgroundDeep Reinforcement Learning in Power Distribution Systems: Overview, Challenges, and Opportunities
Gao, Y., & Yu, N. (2021) · IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT) · doi:10.1109/isgt49243.2021.9372283 · Cited by 36
“Although DRL has achieved great success in many domains such as Atari games [2], data center cooling system control [3], and Go game [4], it has not seen wide-spread adoption in critical infrastructure systems such as power distribution systems because they must be operated in an extremely reliable…”
BackgroundSatô, T., & Fujita, M. (2021) · IEEE Access · doi:10.1109/access.2021.3078668 · Cited by 33
“Especially, the computer player for games of perfect information has achieved super-human performance starting from random play based on the reinforcement learning approach [30], [31], in which a general-purpose Monte Carlo tree search algorithm is effectively employed.”
BackgroundQuantum imaginary time evolution steered by reinforcement learning
Cao, C., An, Z., Hou, S., Zhou, D. L., & Zeng, B. (2022) · Communications Physics · doi:10.1038/s42005-022-00837-y · Cited by 34
“[29] David Silver, Thomas Hubert, Julian Schrittwieser, Ioannis Antonoglou, Matthew Lai, Arthur Guez, Marc Lanctot, Laurent Sifre, Dharshan Kumaran, Thore Graepel, Timothy Lillicrap, Karen Simonyan, and Demis Hassabis, “A general reinforcement learning algorithm that masters chess, shogi, and go through self-play,” Sc…”
BackgroundLearning to traverse over graphs with a Monte Carlo tree search-based self-play framework
Wang, Q., Hao, Y., & Cao, J. (2021) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2021.104422 · Cited by 34
Reinforcement learning for active distribution network planning based on Monte Carlo tree search
Zhang, X., Hua, W., Liu, Y., Duan, J., Tang, Z., & Liu, J. (2021) · International Journal of Electrical Power & Energy Systems · doi:10.1016/j.ijepes.2021.107885 · Cited by 26
Deep reinforcement learning-based long-range autonomous valet parking for smart cities
Khalid, M., Wang, L., Wang, K., Aslam, N., Pan, C., & Cao, Y. (2022) · Sustainable Cities and Society · doi:10.1016/j.scs.2022.104311 · Cited by 30
An Analysis of Body Language of Patients Using Artificial Intelligence
Abdulghafor, R. A. A., Abdelmohsen, A., Turaev, S., Ali, M. A. H., & Wani, S. (2022) · Healthcare · doi:10.3390/healthcare10122504 · Cited by 36
“For example, Google’s DeepMind, which is best known for its AlphaGo game-playing algorithm, AI has been used in [122] to anticipate the structure of viral proteins, which might aid in the development of novel treatments.”
MethodOnline 3D Bin Packing Reinforcement Learning Solution with Buffer
Puche, A. V., & Lee, S. (2022) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros47612.2022.9982095 · Cited by 30
“Regarding model-based methods, the algorithm can either assume the model of the environment is given [12], [13] or either can be learnt [14], [15].”
MethodLog-law recovery through reinforcement-learning wall model for large eddy simulation
Vadrot, A., Yang, X. I. A., Bae, H. J., & Abkar, M. (2023) · Physics of Fluids · doi:10.1063/5.0147570 · Cited by 26
Solving the train dispatching problem via deep reinforcement learning
Agasucci, V., Grani, G., & Lamorgese, L. (2020) · Journal of Rail Transport Planning & Management · doi:10.1016/j.jrtpm.2023.100394 · Cited by 23
“Several approaches have followed AlphaGo, like AlphaZero [33, 32] and MuZero [29], increasing every time the degree of generalization possible.”
MethodA Continuous Action Space Tree search for INverse desiGn (CASTING) framework for materials discovery
Banik, S., Loeffler, T. D., Manna, S., Chan, H., Srinivasan, S., Darancet, P., Hexemer, A., & Sankaranarayanan, S. K. R. S. (2023) · npj Computational Materials · doi:10.1038/s41524-023-01128-y · Cited by 32
“Chemical science 2020, 11(40): 10959-10972.”
Xiao, Y., Yang, Y., Yu, H., & Liu, J. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3379191 · Cited by 32
“DRLhasbeenfeaturedin manygroundbreakingAIapplicationssuchasAlphaZero[23] andChatGPT[24].”
Kirk, J. R., & Laird, J. E. (2019) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2019/844 · Cited by 35
Automated Architecture Design for Deep Neural NetworksPreprint
Abreu, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1908.10714 · Cited by 55
“AlphaZero’s deep neural network coupled with a reinforcement learning algorithm beat the world champion in Go - a game that was previously believed to be too complex to be played competitively by a machine [Silver et al., 2018].”
Uncovering instabilities in variational-quantum deep Q-networks
Franz, M., Wolf, L., Periyasamy, M., Ufrecht, C., Scherer, D. D., Plinge, A., Mutschler, C., & Mauerer, W. (2023) · Journal of the Franklin Institute · doi:10.1016/j.jfranklin.2022.08.021 · Cited by 26
Recent advancements in implantable neural links based on organic synaptic transistors
Biswas, S., Jang, H.-W., Lee, Y., Choi, H., Kim, Y., Kim, H., & Zhu, Y. (2023) · Exploration · doi:10.1002/exp.20220150 · Cited by 28
Strnad, F., Barfuß, W., Donges, J. F., & Heitzig, J. (2019) · Chaos An Interdisciplinary Journal of Nonlinear Science · doi:10.1063/1.5124673 · Cited by 35
Validation of Image-Based Neural Network Controllers through Adaptive Stress Testing
Julian, K. D., Lee, R., & Kochenderfer, M. J. (2020) · IEEE International Conference on Intelligent Transportation Systems (ITSC) · doi:10.1109/itsc45102.2020.9294549 · Cited by 45
“Additional work has shown that neural networks can play chess [2], classify objects [3], and recognize digits [4].”
BackgroundKoutras, D. I., Kapoutsis, A. C., & Kosmatopoulos, E. B. (2020) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2020.3004780 · Cited by 24
“The recent technological advancements in terms of hardware [1] have enabled the development of deep learning architectures [2] capable of dealing with extremely difficult [3] and abstract [4] problems.”
BackgroundA Roadmap for Reaching the Potential of Brain‐Derived Computing
Aimone, J. B. (2020) · Advanced Intelligent Systems · doi:10.1002/aisy.202000191 · Cited by 30
Multi-UAV Cooperative Task Assignment Based on Half Random Q-Learning
Zhu, P., & Fang, X. (2021) · Symmetry · doi:10.3390/sym13122417 · Cited by 27
Automatic Design of PM Motor Using Monte Carlo Tree Search in Conjunction With Topology Optimization
Sato, H., & Igarashi, H. (2022) · IEEE Transactions on Magnetics · doi:10.1109/tmag.2022.3164926 · Cited by 25
“In this article, a novel automatic design method for PM motors is proposed based on a Monte Carlo tree search (MCTS), which has been used for generic gaming artificial intelligence (AI) [5] and field computations [6].”
Method
Metadata
- Authors
- Silver, D., Hubert, T., Schrittwieser, J., Antonoglou, I., Lai, M., Guez, A., Lanctot, M., Sifre, L., Kumaran, D., Graepel, T., Lillicrap, T., Simonyan, K., & Hassabis, D.
- Affiliations
- Google DeepMind (United Kingdom) (GB)
- Corresponding author
- Silver, David
- Hubert, Thomas
- Schrittwieser, Julian
- Hassabis, Demis
- ORCID
- Venue
- Science
- Published
- 2018-12-07
- Volume
- 362
- Issue
- 6419
- Pages
- 1140-1144
- ISSN
- 0036-8075 (ISSN-L)
- 1095-9203
- OpenAlex
- W2902907165
- Cited by
- 3,562
- References
- 30
- Citations by year
- 2026: 257 · 2025: 427 · 2024: 512 · 2023: 514 · 2022: 569 · 2021: 546 · 2020: 462 · 2019: 256 · 2018: 13
- FWCI
- 179.89
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- Science (publishedVersion)
- PubMed (publishedVersion)
- UCL Discovery (University College London) (submittedVersion)
- Funders
- DeepMind
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Artificial Intelligence in Games
- Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- Physical Sciences › Computer Science › Artificial Intelligence › Artificial Intelligence Applications
- Keywords
- AlphaZero, chess, shogi, Go, evaluation function learning, game tree search
- MeSH terms
- Algorithms, Artificial Intelligence, Humans, Reinforcement, Psychology, Software, Video Games
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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.