Video Games on Atari 2600 Tutankham

Metric: Score (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	Score▼	Extra Data	Paper	Date↕	Code
1	Agent57	2354.91	No	Agent57: Outperforming the Atari Human Benchmark	2020-03-30	Code
2	MuZero	491.48	No	Mastering Atari, Go, Chess and Shogi by Planning...	2019-11-19	Code
3	GDI-I3	423.9	No	Generalized Data Distribution Iteration	2022-06-07	-
4	GDI-I3	423.9	No	Generalized Data Distribution Iteration	2022-06-07	-
5	GDI-H3	418.2	No	Generalized Data Distribution Iteration	2022-06-07	-
6	R2D2	395.3	No	-	-	Code
7	MuZero (Res2 Adam)	347.99	No	Online and Offline Reinforcement Learning by Pla...	2021-04-13	Code
8	A2C + SIL	340.5	No	Self-Imitation Learning	2018-06-14	Code
9	QR-DQN-1	297	No	Distributional Reinforcement Learning with Quant...	2017-10-27	Code
10	IQN	293	No	Implicit Quantile Networks for Distributional Re...	2018-06-14	Code
11	IMPALA (deep)	292.11	No	IMPALA: Scalable Distributed Deep-RL with Import...	2018-02-05	Code
12	C51 noop	280	No	A Distributional Perspective on Reinforcement Le...	2017-07-21	Code
13	Ape-X	272.6	No	Distributed Prioritized Experience Replay	2018-03-02	Code
14	NoisyNet-Dueling	269	No	Noisy Networks for Exploration	2017-06-30	Code
15	DreamerV2	264	No	Mastering Atari with Discrete World Models	2020-10-05	Code
16	ASL DDQN	252.9	No	Train a Real-world Local Path Planner in One Hou...	2023-05-07	Code
17	Prior+Duel noop	245.9	No	Dueling Network Architectures for Deep Reinforce...	2015-11-20	Code
18	Advantage Learning	245.22	No	Increasing the Action Gap: New Operators for Rei...	2015-12-15	Code
19	POP3D	241.21	No	Policy Optimization With Penalized Point Probabi...	2018-07-02	Code
20	UCT	225.5	No	The Arcade Learning Environment: An Evaluation P...	2012-07-19	Code
21	DDQN (tuned) noop	218.4	No	Dueling Network Architectures for Deep Reinforce...	2015-11-20	Code
22	Bootstrapped DQN	214.8	No	Deep Exploration via Bootstrapped DQN	2016-02-15	Code
23	Duel noop	211.4	No	Dueling Network Architectures for Deep Reinforce...	2015-11-20	Code
24	Prior noop	204.6	No	Prioritized Experience Replay	2015-11-18	Code
25	DARQN soft	197	No	Deep Attention Recurrent Q-Network	2015-12-05	Code
26	Nature DQN	186.7	No	-	-	Code
27	Recurrent Rational DQN Average	184	No	Adaptive Rational Activations to Boost Deep Rein...	2021-02-18	Code
28	DDQN+Pop-Art noop	183.9	No	Learning values across many orders of magnitude	2016-02-24	-
29	Rational DQN Average	179	No	Adaptive Rational Activations to Boost Deep Rein...	2021-02-18	Code
30	A3C FF hs	156.3	No	Asynchronous Methods for Deep Reinforcement Lear...	2016-02-04	Code
31	A3C LSTM hs	144.2	No	Asynchronous Methods for Deep Reinforcement Lear...	2016-02-04	Code
32	ES FF (1 hour) noop	130.3	No	Evolution Strategies as a Scalable Alternative t...	2017-03-10	Code
33	DNA	127	No	DNA: Proximal Policy Optimization with a Dual Ne...	2022-06-20	Code
34	Gorila	118.5	No	Massively Parallel Methods for Deep Reinforcemen...	2015-07-15	Code
35	Best Learner	114.3	No	The Arcade Learning Environment: An Evaluation P...	2012-07-19	Code
36	Prior+Duel hs	108.6	No	Deep Reinforcement Learning with Double Q-learning	2015-09-22	Code
37	SARSA	98.2	No	-	-	-
38	DDQN (tuned) hs	92.2	No	Deep Reinforcement Learning with Double Q-learning	2015-09-22	Code
39	DQN noop	68.1	No	Deep Reinforcement Learning with Double Q-learning	2015-09-22	Code
40	Prior hs	56.9	No	Prioritized Experience Replay	2015-11-18	Code
41	Duel hs	48	No	Dueling Network Architectures for Deep Reinforce...	2015-11-20	Code
42	DQN hs	45.6	No	Deep Reinforcement Learning with Double Q-learning	2015-09-22	Code
43	A3C FF (1 day) hs	26.1	No	Asynchronous Methods for Deep Reinforcement Lear...	2016-02-04	Code
44	CGP	0	No	-	-	Code

#1Agent57SOTA
2354.91
Score· 2020-03-30
Agent57: Outperforming the Atari Human Benchmark Code
#2MuZeroSOTA
491.48
Score· 2019-11-19
Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model Code
#3GDI-I3
423.9
Score· 2022-06-07
Generalized Data Distribution Iteration
#4GDI-I3
423.9
Score· 2022-06-07
Generalized Data Distribution Iteration
#5GDI-H3
418.2
Score· 2022-06-07
Generalized Data Distribution Iteration
#6R2D2
395.3
Score
No paperCode
#7MuZero (Res2 Adam)
347.99
Score· 2021-04-13
Online and Offline Reinforcement Learning by Planning with a Learned Model Code
#8A2C + SILSOTA
340.5
Score· 2018-06-14
Self-Imitation Learning Code
#9QR-DQN-1SOTA
297
Score· 2017-10-27
Distributional Reinforcement Learning with Quantile Regression Code
#10IQN
293
Score· 2018-06-14
Implicit Quantile Networks for Distributional Reinforcement Learning Code
#11IMPALA (deep)
292.11
Score· 2018-02-05
IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures Code
#12C51 noopSOTA
280
Score· 2017-07-21
A Distributional Perspective on Reinforcement Learning Code
#13Ape-X
272.6
Score· 2018-03-02
Distributed Prioritized Experience Replay Code
#14NoisyNet-DuelingSOTA
269
Score· 2017-06-30
Noisy Networks for Exploration Code
#15DreamerV2
264
Score· 2020-10-05
Mastering Atari with Discrete World Models Code
#16ASL DDQN
252.9
Score· 2023-05-07
Train a Real-world Local Path Planner in One Hour via Partially Decoupled Reinforcement Learning and Vectorized Diversity Code
#17Prior+Duel noopSOTA
245.9
Score· 2015-11-20
Dueling Network Architectures for Deep Reinforcement Learning Code
#18Advantage Learning
245.22
Score· 2015-12-15
Increasing the Action Gap: New Operators for Reinforcement Learning Code
#19POP3D
241.21
Score· 2018-07-02
Policy Optimization With Penalized Point Probability Distance: An Alternative To Proximal Policy Optimization Code
#20UCTSOTA
225.5
Score· 2012-07-19
The Arcade Learning Environment: An Evaluation Platform for General Agents Code
#21DDQN (tuned) noop
218.4
Score· 2015-11-20
Dueling Network Architectures for Deep Reinforcement Learning Code
#22Bootstrapped DQN
214.8
Score· 2016-02-15
Deep Exploration via Bootstrapped DQN Code
#23Duel noop
211.4
Score· 2015-11-20
Dueling Network Architectures for Deep Reinforcement Learning Code
#24Prior noop
204.6
Score· 2015-11-18
Prioritized Experience Replay Code
#25DARQN soft
197
Score· 2015-12-05
Deep Attention Recurrent Q-Network Code
#26Nature DQN
186.7
Score
No paperCode
#27Recurrent Rational DQN Average
184
Score· 2021-02-18
Adaptive Rational Activations to Boost Deep Reinforcement Learning Code
#28DDQN+Pop-Art noop
183.9
Score· 2016-02-24
Learning values across many orders of magnitude
#29Rational DQN Average
179
Score· 2021-02-18
Adaptive Rational Activations to Boost Deep Reinforcement Learning Code
#30A3C FF hs
156.3
Score· 2016-02-04
Asynchronous Methods for Deep Reinforcement Learning Code
#31A3C LSTM hs
144.2
Score· 2016-02-04
Asynchronous Methods for Deep Reinforcement Learning Code
#32ES FF (1 hour) noop
130.3
Score· 2017-03-10
Evolution Strategies as a Scalable Alternative to Reinforcement Learning Code
#33DNA
127
Score· 2022-06-20
DNA: Proximal Policy Optimization with a Dual Network Architecture Code
#34Gorila
118.5
Score· 2015-07-15
Massively Parallel Methods for Deep Reinforcement Learning Code
#35Best Learner
114.3
Score· 2012-07-19
The Arcade Learning Environment: An Evaluation Platform for General Agents Code
#36Prior+Duel hs
108.6
Score· 2015-09-22
Deep Reinforcement Learning with Double Q-learning Code
#37SARSA
98.2
Score
No paper
#38DDQN (tuned) hs
92.2
Score· 2015-09-22
Deep Reinforcement Learning with Double Q-learning Code
#39DQN noop
68.1
Score· 2015-09-22
Deep Reinforcement Learning with Double Q-learning Code
#40Prior hs
56.9
Score· 2015-11-18
Prioritized Experience Replay Code
#41Duel hs
48
Score· 2015-11-20
Dueling Network Architectures for Deep Reinforcement Learning Code
#42DQN hs
45.6
Score· 2015-09-22
Deep Reinforcement Learning with Double Q-learning Code
#43A3C FF (1 day) hs
26.1
Score· 2016-02-04
Asynchronous Methods for Deep Reinforcement Learning Code
#44CGP
0
Score
No paperCode