Tasks
SotA
Datasets
Papers
Methods
Submit
About
SotA
/
Playing Games
/
Video Games
/
Atari 2600 Breakout
Video Games on Atari 2600 Breakout
Metric: Score (higher is better)
Leaderboard
Dataset
Loading chart...
Results
Submit a result
Export CSV
Sort:
Score (best first)
Score (worst first)
Date (newest first)
Date (oldest first)
Model name (A→Z)
#
Model
↕
Score
▼
Extra Data
Paper
Date
↕
Code
1
GDI-H3(200M frames)
864
No
Generalized Data Distribution Iteration
2022-06-07
-
2
GDI-I3(200M frames)
864
No
Generalized Data Distribution Iteration
2022-06-07
-
3
GDI-I3
864
No
Generalized Data Distribution Iteration
2022-06-07
-
4
GDI-H3
864
No
Generalized Data Distribution Iteration
2022-06-07
-
5
Bootstrapped DQN
855
No
Deep Exploration via Bootstrapped DQN
2016-02-15
Code
6
FQF
854.2
No
Fully Parameterized Quantile Function for Distri...
2019-11-05
Code
7
R2D2
837.7
No
-
-
Code
8
Ape-X
800.9
No
Distributed Prioritized Experience Replay
2018-03-02
Code
9
Agent57
790.4
No
Agent57: Outperforming the Atari Human Benchmark
2020-03-30
Code
10
IMPALA (deep)
787.34
No
IMPALA: Scalable Distributed Deep-RL with Import...
2018-02-05
Code
11
A3C LSTM hs
766.8
No
Asynchronous Methods for Deep Reinforcement Lear...
2016-02-04
Code
12
MuZero (Res2 Adam)
758.04
No
Online and Offline Reinforcement Learning by Pla...
2021-04-13
Code
13
C51 noop
748
No
A Distributional Perspective on Reinforcement Le...
2017-07-21
Code
14
QR-DQN-1
742
No
Distributional Reinforcement Learning with Quant...
2017-10-27
Code
15
IQN
734
No
Implicit Quantile Networks for Distributional Re...
2018-06-14
Code
16
A3C FF hs
681.9
No
Asynchronous Methods for Deep Reinforcement Lear...
2016-02-04
Code
17
DNA
626
No
DNA: Proximal Policy Optimization with a Dual Ne...
2022-06-20
Code
18
ASL DDQN
621.7
No
Train a Real-world Local Path Planner in One Hou...
2023-05-07
Code
19
A3C FF (1 day) hs
551.6
No
Asynchronous Methods for Deep Reinforcement Lear...
2016-02-04
Code
20
Reactor 500M
514.8
No
The Reactor: A fast and sample-efficient Actor-C...
2017-04-15
-
21
POP3D
458.41
No
Policy Optimization With Penalized Point Probabi...
2018-07-02
Code
22
A2C + SIL
452
No
Self-Imitation Learning
2018-06-14
Code
23
Persistent AL
431.89
No
Increasing the Action Gap: New Operators for Rei...
2015-12-15
Code
24
Advantage Learning
425.32
No
Increasing the Action Gap: New Operators for Rei...
2015-12-15
Code
25
DDQN (tuned) noop
418.5
No
Dueling Network Architectures for Deep Reinforce...
2015-11-20
Code
26
Duel hs
411.6
No
Dueling Network Architectures for Deep Reinforce...
2015-11-20
Code
27
Nature DQN
401.2
No
-
-
Code
28
DQN noop
385.5
No
Deep Reinforcement Learning with Double Q-learning
2015-09-22
Code
29
Prior noop
373.9
No
Prioritized Experience Replay
2015-11-18
Code
30
MAC
372.7
No
Mean Actor Critic
2017-09-01
Code
31
DDQN (tuned) hs
368.9
No
Deep Reinforcement Learning with Double Q-learning
2015-09-22
Code
32
Prior+Duel noop
366
No
Dueling Network Architectures for Deep Reinforce...
2015-11-20
Code
33
UCT
364.4
No
The Arcade Learning Environment: An Evaluation P...
2012-07-19
Code
34
Prior+Duel hs
354.6
No
Deep Reinforcement Learning with Double Q-learning
2015-09-22
Code
35
DQN hs
354.5
No
Deep Reinforcement Learning with Double Q-learning
2015-09-22
Code
36
DDRL A3C
350
No
Distributed Deep Reinforcement Learning: Learn h...
2018-01-09
Code
37
Duel noop
345.3
No
Dueling Network Architectures for Deep Reinforce...
2015-11-20
Code
38
DDQN+Pop-Art noop
344.1
No
Learning values across many orders of magnitude
2016-02-24
-
39
Prior hs
343
No
Prioritized Experience Replay
2015-11-18
Code
40
Recurrent Rational DQN Average
336
No
Adaptive Rational Activations to Boost Deep Rein...
2021-02-18
Code
41
Rational DQN Average
316
No
Adaptive Rational Activations to Boost Deep Rein...
2021-02-18
Code
42
Gorila
313
No
Massively Parallel Methods for Deep Reinforcemen...
2015-07-15
Code
43
DreamerV2
312
No
Mastering Atari with Discrete World Models
2020-10-05
Code
44
DT
267.5
No
Decision Transformer: Reinforcement Learning via...
2021-06-02
Code
45
NoisyNet-Dueling
263
No
Noisy Networks for Exploration
2017-06-30
Code
46
DQN Best
225
No
Playing Atari with Deep Reinforcement Learning
2013-12-19
Code
47
SPOS
180.6
No
Optimizing the Neural Architecture of Reinforcem...
2020-11-30
Code
48
ENAS Search space 1
161.1
No
Optimizing the Neural Architecture of Reinforcem...
2020-11-30
Code
49
SPOS Search space 1
144.4
No
Optimizing the Neural Architecture of Reinforcem...
2020-11-30
Code
50
ENAS
91.4
No
Optimizing the Neural Architecture of Reinforcem...
2020-11-30
Code
51
DARQN hard
20
No
Deep Attention Recurrent Q-Network
2015-12-05
Code
52
CURL
18.2
No
CURL: Contrastive Unsupervised Representations f...
2020-04-08
Code
53
CGP
13.2
No
Evolving simple programs for playing Atari games
2018-06-14
Code
54
Discrete Latent Space World Model (VQ-VAE)
11.6
No
Smaller World Models for Reinforcement Learning
2020-10-12
-
55
ES FF (1 hour) noop
9.5
No
Evolution Strategies as a Scalable Alternative t...
2017-03-10
Code
56
SARSA
6.1
No
-
-
-
57
Best Learner
5.2
No
The Arcade Learning Environment: An Evaluation P...
2012-07-19
Code
58
SAC
0.7
No
Soft Actor-Critic for Discrete Action Settings
2019-10-16
Code
#1
GDI-H3(200M frames)
SOTA
864
Score
· 2022-06-07
Generalized Data Distribution Iteration
#2
GDI-I3(200M frames)
864
Score
· 2022-06-07
Generalized Data Distribution Iteration
#3
GDI-I3
864
Score
· 2022-06-07
Generalized Data Distribution Iteration
#4
GDI-H3
864
Score
· 2022-06-07
Generalized Data Distribution Iteration
#5
Bootstrapped DQN
SOTA
855
Score
· 2016-02-15
Deep Exploration via Bootstrapped DQN
Code
#6
FQF
854.2
Score
· 2019-11-05
Fully Parameterized Quantile Function for Distributional Reinforcement Learning
Code
#7
R2D2
837.7
Score
No paper
Code
#8
Ape-X
800.9
Score
· 2018-03-02
Distributed Prioritized Experience Replay
Code
#9
Agent57
790.4
Score
· 2020-03-30
Agent57: Outperforming the Atari Human Benchmark
Code
#10
IMPALA (deep)
787.34
Score
· 2018-02-05
IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures
Code
#11
A3C LSTM hs
SOTA
766.8
Score
· 2016-02-04
Asynchronous Methods for Deep Reinforcement Learning
Code
#12
MuZero (Res2 Adam)
758.04
Score
· 2021-04-13
Online and Offline Reinforcement Learning by Planning with a Learned Model
Code
#13
C51 noop
748
Score
· 2017-07-21
A Distributional Perspective on Reinforcement Learning
Code
#14
QR-DQN-1
742
Score
· 2017-10-27
Distributional Reinforcement Learning with Quantile Regression
Code
#15
IQN
734
Score
· 2018-06-14
Implicit Quantile Networks for Distributional Reinforcement Learning
Code
#16
A3C FF hs
681.9
Score
· 2016-02-04
Asynchronous Methods for Deep Reinforcement Learning
Code
#17
DNA
626
Score
· 2022-06-20
DNA: Proximal Policy Optimization with a Dual Network Architecture
Code
#18
ASL DDQN
621.7
Score
· 2023-05-07
Train a Real-world Local Path Planner in One Hour via Partially Decoupled Reinforcement Learning and Vectorized Diversity
Code
#19
A3C FF (1 day) hs
551.6
Score
· 2016-02-04
Asynchronous Methods for Deep Reinforcement Learning
Code
#20
Reactor 500M
514.8
Score
· 2017-04-15
The Reactor: A fast and sample-efficient Actor-Critic agent for Reinforcement Learning
#21
POP3D
458.41
Score
· 2018-07-02
Policy Optimization With Penalized Point Probability Distance: An Alternative To Proximal Policy Optimization
Code
#22
A2C + SIL
452
Score
· 2018-06-14
Self-Imitation Learning
Code
#23
Persistent AL
SOTA
431.89
Score
· 2015-12-15
Increasing the Action Gap: New Operators for Reinforcement Learning
Code
#24
Advantage Learning
425.32
Score
· 2015-12-15
Increasing the Action Gap: New Operators for Reinforcement Learning
Code
#25
DDQN (tuned) noop
SOTA
418.5
Score
· 2015-11-20
Dueling Network Architectures for Deep Reinforcement Learning
Code
#26
Duel hs
411.6
Score
· 2015-11-20
Dueling Network Architectures for Deep Reinforcement Learning
Code
#27
Nature DQN
401.2
Score
No paper
Code
#28
DQN noop
SOTA
385.5
Score
· 2015-09-22
Deep Reinforcement Learning with Double Q-learning
Code
#29
Prior noop
373.9
Score
· 2015-11-18
Prioritized Experience Replay
Code
#30
MAC
372.7
Score
· 2017-09-01
Mean Actor Critic
Code
#31
DDQN (tuned) hs
368.9
Score
· 2015-09-22
Deep Reinforcement Learning with Double Q-learning
Code
#32
Prior+Duel noop
366
Score
· 2015-11-20
Dueling Network Architectures for Deep Reinforcement Learning
Code
#33
UCT
SOTA
364.4
Score
· 2012-07-19
The Arcade Learning Environment: An Evaluation Platform for General Agents
Code
#34
Prior+Duel hs
354.6
Score
· 2015-09-22
Deep Reinforcement Learning with Double Q-learning
Code
#35
DQN hs
354.5
Score
· 2015-09-22
Deep Reinforcement Learning with Double Q-learning
Code
#36
DDRL A3C
350
Score
· 2018-01-09
Distributed Deep Reinforcement Learning: Learn how to play Atari games in 21 minutes
Code
#37
Duel noop
345.3
Score
· 2015-11-20
Dueling Network Architectures for Deep Reinforcement Learning
Code
#38
DDQN+Pop-Art noop
344.1
Score
· 2016-02-24
Learning values across many orders of magnitude
#39
Prior hs
343
Score
· 2015-11-18
Prioritized Experience Replay
Code
#40
Recurrent Rational DQN Average
336
Score
· 2021-02-18
Adaptive Rational Activations to Boost Deep Reinforcement Learning
Code
#41
Rational DQN Average
316
Score
· 2021-02-18
Adaptive Rational Activations to Boost Deep Reinforcement Learning
Code
#42
Gorila
313
Score
· 2015-07-15
Massively Parallel Methods for Deep Reinforcement Learning
Code
#43
DreamerV2
312
Score
· 2020-10-05
Mastering Atari with Discrete World Models
Code
#44
DT
267.5
Score
· 2021-06-02
Decision Transformer: Reinforcement Learning via Sequence Modeling
Code
#45
NoisyNet-Dueling
263
Score
· 2017-06-30
Noisy Networks for Exploration
Code
#46
DQN Best
225
Score
· 2013-12-19
Playing Atari with Deep Reinforcement Learning
Code
#47
SPOS
180.6
Score
· 2020-11-30
Optimizing the Neural Architecture of Reinforcement Learning Agents
Code
#48
ENAS Search space 1
161.1
Score
· 2020-11-30
Optimizing the Neural Architecture of Reinforcement Learning Agents
Code
#49
SPOS Search space 1
144.4
Score
· 2020-11-30
Optimizing the Neural Architecture of Reinforcement Learning Agents
Code
#50
ENAS
91.4
Score
· 2020-11-30
Optimizing the Neural Architecture of Reinforcement Learning Agents
Code
#51
DARQN hard
20
Score
· 2015-12-05
Deep Attention Recurrent Q-Network
Code
#52
CURL
18.2
Score
· 2020-04-08
CURL: Contrastive Unsupervised Representations for Reinforcement Learning
Code
#53
CGP
13.2
Score
· 2018-06-14
Evolving simple programs for playing Atari games
Code
#54
Discrete Latent Space World Model (VQ-VAE)
11.6
Score
· 2020-10-12
Smaller World Models for Reinforcement Learning
#55
ES FF (1 hour) noop
9.5
Score
· 2017-03-10
Evolution Strategies as a Scalable Alternative to Reinforcement Learning
Code
#56
SARSA
6.1
Score
No paper
#57
Best Learner
5.2
Score
· 2012-07-19
The Arcade Learning Environment: An Evaluation Platform for General Agents
Code
#58
SAC
0.7
Score
· 2019-10-16
Soft Actor-Critic for Discrete Action Settings
Code