List your company

ResearchFormerly WizWand

LunarLander

Benchmarks

TaskDatasetSOTA resultTrendResults

Reinforcement Learning

LunarLander v2

2,292Final Return

30

Reinforcement Learning

LunarLander

283.56Average Episode Reward

15

Reinforcement Learning

LunarLander v3

289Average Agent Reward

14

Continuous-state and discrete-action control

LunarLander v3

232.9Average Reward

13

Reinforcement Learning

LunarLander

0.8Training Time (min)

13

LunarLander

0.18Robustness Gap

12

Black-box Optimization

LunarLander frozen noise v3

0Total Proposal Time (s)

9

Reinforcement Learning

LunarLander

260.6Maximum Return

9

Continuous Control

LunarLander Nonmarkov v2 (test)

107.6AUC@T

9

Lunarlander

0.45Regret

8

Black-box Optimization

LunarLander natural noise v3

0.3Total Proposal Time (s)

8

Reinforcement Learning

LunarLander v3

3.2Coefficient of Variation

8

Reinforcement Learning

LunarLander v3

290Episodes to Threshold (Score 200)

8

Reinforcement Learning

LunarLander classical control 1M steps

267.19Return

8

Offline Reinforcement Learning

LunarLander Gym (100k)

107Returns

6

Offline Reinforcement Learning

LunarLander Gym (10k)

102Returns

6

Offline Reinforcement Learning

LunarLander Gym (1k)

97Returns

6

Reinforcement Learning

LunarLander (LL) (test)

241Average Undiscounted Reward

6

Trajectory Ranking

LunarLander v2

207.13Average Reward

6

Reinforcement Learning Surrogate Modeling

LunarLander (LL) (test)

90Reward Ratio

4

Reinforcement Learning

LunarLander v3

11,820,726Training Steps

4

Reinforcement Learning

LunarLander Continuous v3

6.6Arrival Time (T=200)

4

Reinforcement Learning

LunarLander

166.3Return

3

Reinforcement Learning

LunarLander

400,000Environment Episodes

3

Meta-Reinforcement Learning

Lunarlander g

0.015FLOPs (k)

3