List your company

ResearchFormerly WizWand

Walker2D

Benchmarks

TaskDatasetSOTA resultTrendResults

Walker2D

1,023Reward

16

Offline Policy Adaptation

walker2d medium-expert

62.1Normalized Score

14

Offline Policy Adaptation

walker2d medium-replay

63.1Normalized Score

14

Offline Policy Adaptation

walker2d medium

56.7Normalized Score

14

Offline Reinforcement Learning

Walker2d random

21.5Normalized Score

14

Walker2D

4,411Reward

12

Continuous Control

Walker2d 6-Dof

5,143Final Return

12

Offline Reinforcement Learning

Walker2d Expert

211.31Episodic Return

12

Imitation Learning

Walker2d one-shot v2

70Normalized Score

11

Walker2d v4

5,472.8Mean Return

10

Continuous Control

Walker2d v3

4,013Average Return

10

Trojan Attack (Target action: 'fixed random')

Walker2d Walk

98.5Attack Success Rate (ASR)

9

Trojan Attack (Target action: 'arithmetic')

Walker2d Walk

100ASR

9

Trojan Attack (Target action: '1')

Walker2d Walk

100ASR

9

walker2d medium-expert v2

5,014.9Mean Raw MuJoCo Return

9

walker2d medium-expert v2

4,919.8Mean Raw MuJoCo Return

9

walker2d medium-replay v2

581.5Mean Raw MuJoCo Return

9

walker2d medium-replay v2

581.5Mean Raw Return

9

walker2d medium-replay v2

581.5Mean Raw MuJoCo Return

9

Offline Reinforcement Learning

Walker2D Medium-Expert (BodyMass shift)

94.578Performance Score

9

Offline Reinforcement Learning

Walker2D Medium JointNoise shift

80.904Performance Score

9

Offline Reinforcement Learning

Walker2D Medium-Expert JointNoise Shift

118.564Average Return

9

Offline Reinforcement Learning

Walker2D Medium-Replay JointNoise Shift

82.363Average Return

9

Offline Reinforcement Learning

Walker2D JointNoise Shift (Medium)

80.904Average Return

9

Continuous Control

walker2d

5,111Avg Reward

9