Human-level control through deep reinforcement learning

Subjects

Abstract

The theory of reinforcement learning provides a normative account1, deeply rooted in psychological2 and neuroscientific3 perspectives on animal behaviour, of how agents may optimize their control of an environment. To use reinforcement learning successfully in situations approaching real-world complexity, however, agents are confronted with a difficult task: they must derive efficient representations of the environment from high-dimensional sensory inputs, and use these to generalize past experience to new situations. Remarkably, humans and other animals seem to solve this problem through a harmonious combination of reinforcement learning and hierarchical sensory processing systems4,5, the former evidenced by a wealth of neural data revealing notable parallels between the phasic signals emitted by dopaminergic neurons and temporal difference reinforcement learning algorithms3. While reinforcement learning agents have achieved some successes in a variety of domains6,7,8, their applicability has previously been limited to domains in which useful features can be handcrafted, or to domains with fully observed, low-dimensional state spaces. Here we use recent advances in training deep neural networks9,10,11 to develop a novel artificial agent, termed a deep Q-network, that can learn successful policies directly from high-dimensional sensory inputs using end-to-end reinforcement learning. We tested this agent on the challenging domain of classic Atari 2600 games12. We demonstrate that the deep Q-network agent, receiving only the pixels and the game score as inputs, was able to surpass the performance of all previous algorithms and achieve a level comparable to that of a professional human games tester across a set of 49 games, using the same algorithm, network architecture and hyperparameters. This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks.

Access options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Figure 1: Schematic illustration of the convolutional neural network.
Figure 2: Training curves tracking the agent’s average score and average predicted action-value.
Figure 3: Comparison of the DQN agent with the best reinforcement learning methods15 in the literature.
Figure 4: Two-dimensional t-SNE embedding of the representations in the last hidden layer assigned by DQN to game states experienced while playing Space Invaders.

References

  1. 1

    Sutton, R. & Barto, A. Reinforcement Learning: An Introduction (MIT Press, 1998)

    Google Scholar 

  2. 2

    Thorndike, E. L. Animal Intelligence: Experimental studies (Macmillan, 1911)

    Google Scholar 

  3. 3

    Schultz, W., Dayan, P. & Montague, P. R. A neural substrate of prediction and reward. Science 275, 1593–1599 (1997)

    CAS  Article  Google Scholar 

  4. 4

    Serre, T., Wolf, L. & Poggio, T. Object recognition with features inspired by visual cortex. Proc. IEEE. Comput. Soc. Conf. Comput. Vis. Pattern. Recognit. 994–1000 (2005)

    Google Scholar 

  5. 5

    Fukushima, K. Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biol. Cybern. 36, 193–202 (1980)

    CAS  Article  Google Scholar 

  6. 6

    Tesauro, G. Temporal difference learning and TD-Gammon. Commun. ACM 38, 58–68 (1995)

    Article  Google Scholar 

  7. 7

    Riedmiller, M., Gabel, T., Hafner, R. & Lange, S. Reinforcement learning for robot soccer. Auton. Robots 27, 55–73 (2009)

    Article  Google Scholar 

  8. 8

    Diuk, C., Cohen, A. & Littman, M. L. An object-oriented representation for efficient reinforcement learning. Proc. Int. Conf. Mach. Learn. 240–247 (2008)

    Google Scholar 

  9. 9

    Bengio, Y. Learning deep architectures for AI. Foundations and Trends in Machine Learning 2, 1–127 (2009)

    Article  Google Scholar 

  10. 10

    Krizhevsky, A., Sutskever, I. & Hinton, G. ImageNet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst. 25, 1106–1114 (2012)

    Google Scholar 

  11. 11

    Hinton, G. E. & Salakhutdinov, R. R. Reducing the dimensionality of data with neural networks. Science 313, 504–507 (2006)

    CAS  ADS  MathSciNet  Article  Google Scholar 

  12. 12

    Bellemare, M. G., Naddaf, Y., Veness, J. & Bowling, M. The arcade learning environment: An evaluation platform for general agents. J. Artif. Intell. Res. 47, 253–279 (2013)

    Article  Google Scholar 

  13. 13

    Legg, S. & Hutter, M. Universal Intelligence: a definition of machine intelligence. Minds Mach. 17, 391–444 (2007)

    Article  Google Scholar 

  14. 14

    Genesereth, M., Love, N. & Pell, B. General game playing: overview of the AAAI competition. AI Mag. 26, 62–72 (2005)

    Google Scholar 

  15. 15

    Bellemare, M. G., Veness, J. & Bowling, M. Investigating contingency awareness using Atari 2600 games. Proc. Conf. AAAI. Artif. Intell. 864–871 (2012)

    Google Scholar 

  16. 16

    McClelland, J. L., Rumelhart, D. E. & Group, T. P. R. Parallel Distributed Processing: Explorations in the Microstructure of Cognition (MIT Press, 1986)

    Google Scholar 

  17. 17

    LeCun, Y., Bottou, L., Bengio, Y. & Haffner, P. Gradient-based learning applied to document recognition. Proc. IEEE 86, 2278–2324 (1998)

    Article  Google Scholar 

  18. 18

    Hubel, D. H. & Wiesel, T. N. Shape and arrangement of columns in cat’s striate cortex. J. Physiol. 165, 559–568 (1963)

    CAS  Article  Google Scholar 

  19. 19

    Watkins, C. J. & Dayan, P. Q-learning. Mach. Learn. 8, 279–292 (1992)

    MATH  Google Scholar 

  20. 20

    Tsitsiklis, J. & Roy, B. V. An analysis of temporal-difference learning with function approximation. IEEE Trans. Automat. Contr. 42, 674–690 (1997)

    MathSciNet  Article  Google Scholar 

  21. 21

    McClelland, J. L., McNaughton, B. L. & O’Reilly, R. C. Why there are complementary learning systems in the hippocampus and neocortex: insights from the successes and failures of connectionist models of learning and memory. Psychol. Rev. 102, 419–457 (1995)

    Article  Google Scholar 

  22. 22

    O’Neill, J., Pleydell-Bouverie, B., Dupret, D. & Csicsvari, J. Play it again: reactivation of waking experience and memory. Trends Neurosci. 33, 220–229 (2010)

    Article  Google Scholar 

  23. 23

    Lin, L.-J. Reinforcement learning for robots using neural networks. Technical Report, DTIC Document. (1993)

  24. 24

    Riedmiller, M. Neural fitted Q iteration - first experiences with a data efficient neural reinforcement learning method. Mach. Learn.: ECML 3720, 317–328 (Springer, 2005)

    Google Scholar 

  25. 25

    Van der Maaten, L. J. P. & Hinton, G. E. Visualizing high-dimensional data using t-SNE. J. Mach. Learn. Res. 9, 2579–2605 (2008)

    MATH  Google Scholar 

  26. 26

    Lange, S. & Riedmiller, M. Deep auto-encoder neural networks in reinforcement learning. Proc. Int. Jt. Conf. Neural. Netw. 1–8 (2010)

    Google Scholar 

  27. 27

    Law, C.-T. & Gold, J. I. Reinforcement learning can account for associative and perceptual learning on a visual decision task. Nature Neurosci. 12, 655 (2009)

    CAS  Article  Google Scholar 

  28. 28

    Sigala, N. & Logothetis, N. K. Visual categorization shapes feature selectivity in the primate temporal cortex. Nature 415, 318–320 (2002)

    CAS  ADS  Article  Google Scholar 

  29. 29

    Bendor, D. & Wilson, M. A. Biasing the content of hippocampal replay during sleep. Nature Neurosci. 15, 1439–1444 (2012)

    CAS  Article  Google Scholar 

  30. 30

    Moore, A. & Atkeson, C. Prioritized sweeping: reinforcement learning with less data and less real time. Mach. Learn. 13, 103–130 (1993)

    Google Scholar 

  31. 31

    Jarrett, K., Kavukcuoglu, K., Ranzato, M. A. & LeCun, Y. What is the best multi-stage architecture for object recognition? Proc. IEEE. Int. Conf. Comput. Vis. 2146–2153 (2009)

    Google Scholar 

  32. 32

    Nair, V. & Hinton, G. E. Rectified linear units improve restricted Boltzmann machines. Proc. Int. Conf. Mach. Learn. 807–814 (2010)

  33. 33

    Kaelbling, L. P., Littman, M. L. & Cassandra, A. R. Planning and acting in partially observable stochastic domains. Artificial Intelligence 101, 99–134 (1994)

    MathSciNet  Article  Google Scholar 

Download references

Acknowledgements

We thank G. Hinton, P. Dayan and M. Bowling for discussions, A. Cain and J. Keene for work on the visuals, K. Keller and P. Rogers for help with the visuals, G. Wayne for comments on an earlier version of the manuscript, and the rest of the DeepMind team for their support, ideas and encouragement.

Author information

Affiliations

Authors

Contributions

V.M., K.K., D.S., J.V., M.G.B., M.R., A.G., D.W., S.L. and D.H. conceptualized the problem and the technical framework. V.M., K.K., A.A.R. and D.S. developed and tested the algorithms. J.V., S.P., C.B., A.A.R., M.G.B., I.A., A.K.F., G.O. and A.S. created the testing platform. K.K., H.K., S.L. and D.H. managed the project. K.K., D.K., D.H., V.M., D.S., A.G., A.A.R., J.V. and M.G.B. wrote the paper.

Corresponding authors

Correspondence to Koray Kavukcuoglu or Demis Hassabis.

Ethics declarations

Competing interests

The authors declare no competing financial interests.

Extended data figures and tables

Extended Data Figure 1 Two-dimensional t-SNE embedding of the representations in the last hidden layer assigned by DQN to game states experienced during a combination of human and agent play in Space Invaders.

The plot was generated by running the t-SNE algorithm25 on the last hidden layer representation assigned by DQN to game states experienced during a combination of human (30 min) and agent (2 h) play. The fact that there is similar structure in the two-dimensional embeddings corresponding to the DQN representation of states experienced during human play (orange points) and DQN play (blue points) suggests that the representations learned by DQN do indeed generalize to data generated from policies other than its own. The presence in the t-SNE embedding of overlapping clusters of points corresponding to the network representation of states experienced during human and agent play shows that the DQN agent also follows sequences of states similar to those found in human play. Screenshots corresponding to selected states are shown (human: orange border; DQN: blue border).

Extended Data Figure 2 Visualization of learned value functions on two games, Breakout and Pong.

a, A visualization of the learned value function on the game Breakout. At time points 1 and 2, the state value is predicted to be 17 and the agent is clearing the bricks at the lowest level. Each of the peaks in the value function curve corresponds to a reward obtained by clearing a brick. At time point 3, the agent is about to break through to the top level of bricks and the value increases to 21 in anticipation of breaking out and clearing a large set of bricks. At point 4, the value is above 23 and the agent has broken through. After this point, the ball will bounce at the upper part of the bricks clearing many of them by itself. b, A visualization of the learned action-value function on the game Pong. At time point 1, the ball is moving towards the paddle controlled by the agent on the right side of the screen and the values of all actions are around 0.7, reflecting the expected value of this state based on previous experience. At time point 2, the agent starts moving the paddle towards the ball and the value of the ‘up’ action stays high while the value of the ‘down’ action falls to −0.9. This reflects the fact that pressing ‘down’ would lead to the agent losing the ball and incurring a reward of −1. At time point 3, the agent hits the ball by pressing ‘up’ and the expected reward keeps increasing until time point 4, when the ball reaches the left edge of the screen and the value of all actions reflects that the agent is about to receive a reward of 1. Note, the dashed line shows the past trajectory of the ball purely for illustrative purposes (that is, not shown during the game). With permission from Atari Interactive, Inc.

Extended Data Table 1 List of hyperparameters and their values
Extended Data Table 2 Comparison of games scores obtained by DQN agents with methods from the literature12,15 and a professional human games tester
Extended Data Table 3 The effects of replay and separating the target Q-network
Extended Data Table 4 Comparison of DQN performance with linear function approximator

Supplementary information

Supplementary Information

This file contains a Supplementary Discussion. (PDF 110 kb)

Performance of DQN in the Game Space Invaders

This video shows the performance of the DQN agent while playing the game of Space Invaders. The DQN agent successfully clears the enemy ships on the screen while the enemy ships move down and sideways with gradually increasing speed. (MOV 5106 kb)

Demonstration of Learning Progress in the Game Breakout

This video shows the improvement in the performance of DQN over training (i.e. after 100, 200, 400 and 600 episodes). After 600 episodes DQN finds and exploits the optimal strategy in this game, which is to make a tunnel around the side, and then allow the ball to hit blocks by bouncing behind the wall. Note: the score is displayed at the top left of the screen (maximum for clearing one screen is 448 points), number of lives remaining is shown in the middle (starting with 5 lives), and the “1” on the top right indicates this is a 1-player game. (MOV 1500 kb)

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Mnih, V., Kavukcuoglu, K., Silver, D. et al. Human-level control through deep reinforcement learning. Nature 518, 529–533 (2015). https://doi.org/10.1038/nature14236

Download citation

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Search

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing