List your company

ResearchFormerly WizWand

BipedalWalker

Benchmarks

TaskDatasetSOTA resultTrendResults

Reinforcement Learning

BipedalWalker

325.35Average Episode Reward

26

Continuous Control

BipedalWalker v3

298.4Episodic Cumulative Reward

15

BipedalWalker Overall Mean

89.95Mean Return

11

BipedalWalker Roughness terrain

224.4Mean Return

11

BipedalWalker Stump terrain

34.16Mean Return

11

BipedalWalker PitGap terrain

-7.65Mean Return

11

BipedalWalker Stairs terrain

-0.66Mean Return

11

BipedalWalker Hardcore terrain

86.83Mean Return

11

BipedalWalker Basic terrain

293.67Mean Return

11

Black-box Optimization

BipedalWalker frozen noise v3

0.3Total Proposal Time (s)

9

BipedalWalker Zero-Shot (test)

100Basic Solved Rate

9

Reinforcement Learning

BipedalWalker

18.38Training Time (h)

9

Continuous Control

BipedalWalker Nonmarkov v3

184.7AUC@T

9

Black-box Optimization

BipedalWalker natural noise v3

1.9Total Proposal Time (s)

8

Robotic Control

BipedalWalker v3

83.5Local Optima Escape Rate

7

Environment Interaction

BipedalWalker

347Environment Steps (M)

7

Adaptability Evaluation

BipedalWalker mass variations (test)

295.65AUC

6

Adaptability Evaluation

BipedalWalker friction variations (test)

1,429.66AUC

6

Quality-Diversity

BipedalWalker

6.09GT QD Score

6

Reinforcement Learning

BipedalWalker v3

273.2Return

6

Critical Scenario Attribution

BipedalWalker

87CTR

4

Reinforcement Learning

bipedalwalker Sticky

42,687,915.83AUC@T

2

Reinforcement Learning

bipedalwalker Noisy

32,301,685.83AUC@T

2

Reinforcement Learning

bipedalwalker (Clean)

9,392,950.11AUC@T

2

Reinforcement Learning

BipedalWalker standard (test)

17Length

2