Atari Games on Atari 2600 Freeway

Metric: Score (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	Score▼	Extra Data	Paper	Date↕	Code
1	TRPO-hash	34	No	#Exploration: A Study of Count-Based Exploration...	2016-11-15	Code
2	IQN	34	No	Implicit Quantile Networks for Distributional Re...	2018-06-14	Code
3	NoisyNet-Dueling	34	No	Noisy Networks for Exploration	2017-06-30	Code
4	QR-DQN-1	34	No	Distributional Reinforcement Learning with Quant...	2017-10-27	Code
5	Go-Explore	34	No	First return, then explore	2020-04-27	Code
6	GDI-I3	34	No	GDI: Rethinking What Makes Reinforcement Learnin...	2021-06-11	-
7	GDI-I3	34	No	GDI: Rethinking What Makes Reinforcement Learnin...	2021-06-11	-
8	GDI-H3(200M frames)	34	No	Generalized Data Distribution Iteration	2022-06-07	-
9	GDI-H3	34	No	Generalized Data Distribution Iteration	2022-06-07	-
10	C51 noop	33.9	No	A Distributional Perspective on Reinforcement Le...	2017-07-21	Code
11	Bootstrapped DQN	33.9	No	Deep Exploration via Bootstrapped DQN	2016-02-15	Code
12	ASL DDQN	33.9	No	Train a Real-world Local Path Planner in One Hou...	2023-05-07	Code
13	MuZero (Res2 Adam)	33.87	No	Online and Offline Reinforcement Learning by Pla...	2021-04-13	Code
14	Prior noop	33.7	No	Prioritized Experience Replay	2015-11-18	Code
15	Ape-X	33.7	No	Distributed Prioritized Experience Replay	2018-03-02	Code
16	DDQN+Pop-Art noop	33.4	No	Learning values across many orders of magnitude	2016-02-24	-
17	DDQN (tuned) noop	33.3	No	Dueling Network Architectures for Deep Reinforce...	2015-11-20	Code
18	MuZero	33.03	No	Mastering Atari, Go, Chess and Shogi by Planning...	2019-11-19	Code
19	Prior+Duel noop	33	No	Dueling Network Architectures for Deep Reinforce...	2015-11-20	Code
20	DQN-CTS	33	No	Count-Based Exploration with Neural Density Models	2017-03-03	Code
21	DreamerV2	33	No	Mastering Atari with Discrete World Models	2020-10-05	Code
22	DNA	33	No	DNA: Proximal Policy Optimization with a Dual Ne...	2022-06-20	Code
23	Intrinsic Reward Agent	32.8	No	Large-Scale Study of Curiosity-Driven Learning	2018-08-13	Code
24	Agent57	32.59	No	Agent57: Outperforming the Atari Human Benchmark	2020-03-30	Code
25	R2D2	32.5	No	-	-	Code
26	Persistent AL	32.3	No	Increasing the Action Gap: New Operators for Rei...	2015-12-15	Code
27	A2C + SIL	32.2	No	Self-Imitation Learning	2018-06-14	Code
28	Advantage Learning	31.72	No	Increasing the Action Gap: New Operators for Rei...	2015-12-15	Code
29	DQN-PixelCNN	31.7	No	Count-Based Exploration with Neural Density Models	2017-03-03	Code
30	ES FF (1 hour) noop	31	No	Evolution Strategies as a Scalable Alternative t...	2017-03-10	Code
31	DQN noop	30.8	No	Deep Reinforcement Learning with Double Q-learning	2015-09-22	Code
32	A3C-CTS	30.48	No	Unifying Count-Based Exploration and Intrinsic M...	2016-06-06	Code
33	Nature DQN	30.3	No	-	-	Code
34	Sarsa-ε	29.9	No	Count-Based Exploration in Feature Space for Rei...	2017-06-25	Code
35	DQNMMCe	29.5	No	Count-Based Exploration with the Successor Repre...	2018-07-31	Code
36	Discrete Latent Space World Model (VQ-VAE)	29	No	Smaller World Models for Reinforcement Learning	2020-10-12	-
37	Prior hs	28.9	No	Prioritized Experience Replay	2015-11-18	Code
38	DDQN (tuned) hs	28.8	No	Deep Reinforcement Learning with Double Q-learning	2015-09-22	Code
39	Prior+Duel hs	28.2	No	Deep Reinforcement Learning with Double Q-learning	2015-09-22	Code
40	CGP	28.2	No	Evolving simple programs for playing Atari games	2018-06-14	Code
41	CURL	27.9	No	CURL: Contrastive Unsupervised Representations f...	2020-04-08	Code
42	MP-EB	27	No	Incentivizing Exploration In Reinforcement Learn...	2015-07-03	Code
43	DQN hs	26.9	No	Deep Reinforcement Learning with Double Q-learning	2015-09-22	Code
44	Best Baseline	22.5	No	The Arcade Learning Environment: An Evaluation P...	2012-07-19	Code
45	ENAS	22	No	Optimizing the Neural Architecture of Reinforcem...	2020-11-30	Code
46	SPOS	22	No	Optimizing the Neural Architecture of Reinforcem...	2020-11-30	Code
47	POP3D	21.21	No	Policy Optimization With Penalized Point Probabi...	2018-07-02	Code
48	SARSA	19.7	No	-	-	-
49	Best Learner	19.1	No	The Arcade Learning Environment: An Evaluation P...	2012-07-19	Code
50	Gorila	10.2	No	Massively Parallel Methods for Deep Reinforcemen...	2015-07-15	Code
51	SAC	4.4	No	Soft Actor-Critic for Discrete Action Settings	2019-10-16	Code
52	UCT	0.4	No	The Arcade Learning Environment: An Evaluation P...	2012-07-19	Code
53	Duel hs	0.2	No	Dueling Network Architectures for Deep Reinforce...	2015-11-20	Code
54	A3C FF (1 day) hs	0.1	No	Asynchronous Methods for Deep Reinforcement Lear...	2016-02-04	Code
55	A3C FF hs	0.1	No	Asynchronous Methods for Deep Reinforcement Lear...	2016-02-04	Code
56	A3C LSTM hs	0.1	No	Asynchronous Methods for Deep Reinforcement Lear...	2016-02-04	Code
57	Duel noop	0	No	-	-	Code
58	Sarsa-φ-EB	0	No	-	-	Code
59	IMPALA (deep)	0	No	-	-	Code

#1TRPO-hashSOTA
34
Score· 2016-11-15
#Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning Code
#2IQN
34
Score· 2018-06-14
Implicit Quantile Networks for Distributional Reinforcement Learning Code
#3NoisyNet-Dueling
34
Score· 2017-06-30
Noisy Networks for Exploration Code
#4QR-DQN-1
34
Score· 2017-10-27
Distributional Reinforcement Learning with Quantile Regression Code
#5Go-Explore
34
Score· 2020-04-27
First return, then explore Code
#6GDI-I3
34
Score· 2021-06-11
GDI: Rethinking What Makes Reinforcement Learning Different From Supervised Learning
#7GDI-I3
34
Score· 2021-06-11
GDI: Rethinking What Makes Reinforcement Learning Different From Supervised Learning
#8GDI-H3(200M frames)
34
Score· 2022-06-07
Generalized Data Distribution Iteration
#9GDI-H3
34
Score· 2022-06-07
Generalized Data Distribution Iteration
#10C51 noop
33.9
Score· 2017-07-21
A Distributional Perspective on Reinforcement Learning Code
#11Bootstrapped DQNSOTA
33.9
Score· 2016-02-15
Deep Exploration via Bootstrapped DQN Code
#12ASL DDQN
33.9
Score· 2023-05-07
Train a Real-world Local Path Planner in One Hour via Partially Decoupled Reinforcement Learning and Vectorized Diversity Code
#13MuZero (Res2 Adam)
33.87
Score· 2021-04-13
Online and Offline Reinforcement Learning by Planning with a Learned Model Code
#14Prior noopSOTA
33.7
Score· 2015-11-18
Prioritized Experience Replay Code
#15Ape-X
33.7
Score· 2018-03-02
Distributed Prioritized Experience Replay Code
#16DDQN+Pop-Art noop
33.4
Score· 2016-02-24
Learning values across many orders of magnitude
#17DDQN (tuned) noop
33.3
Score· 2015-11-20
Dueling Network Architectures for Deep Reinforcement Learning Code
#18MuZero
33.03
Score· 2019-11-19
Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model Code
#19Prior+Duel noop
33
Score· 2015-11-20
Dueling Network Architectures for Deep Reinforcement Learning Code
#20DQN-CTS
33
Score· 2017-03-03
Count-Based Exploration with Neural Density Models Code
#21DreamerV2
33
Score· 2020-10-05
Mastering Atari with Discrete World Models Code
#22DNA
33
Score· 2022-06-20
DNA: Proximal Policy Optimization with a Dual Network Architecture Code
#23Intrinsic Reward Agent
32.8
Score· 2018-08-13
Large-Scale Study of Curiosity-Driven Learning Code
#24Agent57
32.59
Score· 2020-03-30
Agent57: Outperforming the Atari Human Benchmark Code
#25R2D2
32.5
Score
No paperCode
#26Persistent AL
32.3
Score· 2015-12-15
Increasing the Action Gap: New Operators for Reinforcement Learning Code
#27A2C + SIL
32.2
Score· 2018-06-14
Self-Imitation Learning Code
#28Advantage Learning
31.72
Score· 2015-12-15
Increasing the Action Gap: New Operators for Reinforcement Learning Code
#29DQN-PixelCNN
31.7
Score· 2017-03-03
Count-Based Exploration with Neural Density Models Code
#30ES FF (1 hour) noop
31
Score· 2017-03-10
Evolution Strategies as a Scalable Alternative to Reinforcement Learning Code
#31DQN noopSOTA
30.8
Score· 2015-09-22
Deep Reinforcement Learning with Double Q-learning Code
#32A3C-CTS
30.48
Score· 2016-06-06
Unifying Count-Based Exploration and Intrinsic Motivation Code
#33Nature DQN
30.3
Score
No paperCode
#34Sarsa-ε
29.9
Score· 2017-06-25
Count-Based Exploration in Feature Space for Reinforcement Learning Code
#35DQNMMCe
29.5
Score· 2018-07-31
Count-Based Exploration with the Successor Representation Code
#36Discrete Latent Space World Model (VQ-VAE)
29
Score· 2020-10-12
Smaller World Models for Reinforcement Learning
#37Prior hs
28.9
Score· 2015-11-18
Prioritized Experience Replay Code
#38DDQN (tuned) hs
28.8
Score· 2015-09-22
Deep Reinforcement Learning with Double Q-learning Code
#39Prior+Duel hs
28.2
Score· 2015-09-22
Deep Reinforcement Learning with Double Q-learning Code
#40CGP
28.2
Score· 2018-06-14
Evolving simple programs for playing Atari games Code
#41CURL
27.9
Score· 2020-04-08
CURL: Contrastive Unsupervised Representations for Reinforcement Learning Code
#42MP-EBSOTA
27
Score· 2015-07-03
Incentivizing Exploration In Reinforcement Learning With Deep Predictive Models Code
#43DQN hs
26.9
Score· 2015-09-22
Deep Reinforcement Learning with Double Q-learning Code
#44Best BaselineSOTA
22.5
Score· 2012-07-19
The Arcade Learning Environment: An Evaluation Platform for General Agents Code
#45ENAS
22
Score· 2020-11-30
Optimizing the Neural Architecture of Reinforcement Learning Agents Code
#46SPOS
22
Score· 2020-11-30
Optimizing the Neural Architecture of Reinforcement Learning Agents Code
#47POP3D
21.21
Score· 2018-07-02
Policy Optimization With Penalized Point Probability Distance: An Alternative To Proximal Policy Optimization Code
#48SARSA
19.7
Score
No paper
#49Best Learner
19.1
Score· 2012-07-19
The Arcade Learning Environment: An Evaluation Platform for General Agents Code
#50Gorila
10.2
Score· 2015-07-15
Massively Parallel Methods for Deep Reinforcement Learning Code
#51SAC
4.4
Score· 2019-10-16
Soft Actor-Critic for Discrete Action Settings Code
#52UCT
0.4
Score· 2012-07-19
The Arcade Learning Environment: An Evaluation Platform for General Agents Code
#53Duel hs
0.2
Score· 2015-11-20
Dueling Network Architectures for Deep Reinforcement Learning Code
#54A3C FF (1 day) hs
0.1
Score· 2016-02-04
Asynchronous Methods for Deep Reinforcement Learning Code
#55A3C FF hs
0.1
Score· 2016-02-04
Asynchronous Methods for Deep Reinforcement Learning Code
#56A3C LSTM hs
0.1
Score· 2016-02-04
Asynchronous Methods for Deep Reinforcement Learning Code
#57Duel noop
0
Score
No paperCode
#58Sarsa-φ-EB
0
Score
No paperCode
#59IMPALA (deep)
0
Score
No paperCode