BipedalWalker

Benchmarks

Task Name	Dataset Name	SOTA Result
Reinforcement Learning	BipedalWalker	Average Episode Reward314.24	20
Continuous Control	BipedalWalker v3	Episodic Cumulative Reward298.4	15
Locomotion	BipedalWalker Overall Mean	Mean Return89.95	11
Locomotion	BipedalWalker Roughness terrain	Mean Return224.4	11
Locomotion	BipedalWalker Stump terrain	Mean Return34.16	11
Locomotion	BipedalWalker PitGap terrain	Mean Return-7.65	11
Locomotion	BipedalWalker Stairs terrain	Mean Return-0.66	11
Locomotion	BipedalWalker Hardcore terrain	Mean Return86.83	11
Locomotion	BipedalWalker Basic terrain	Mean Return293.67	11
Solved Rate	BipedalWalker Zero-Shot (test)	Basic Solved Rate100	9
Reinforcement Learning	BipedalWalker	Training Time (h)18.38	9
Continuous Control	BipedalWalker Nonmarkov v3	AUC@T184.7	9
Robotic Control	BipedalWalker v3	Local Optima Escape Rate83.5	7
Environment Interaction	BipedalWalker	Environment Steps (M)347	7
Adaptability Evaluation	BipedalWalker mass variations (test)	AUC295.65	6
Adaptability Evaluation	BipedalWalker friction variations (test)	AUC1,429.66	6
Quality-Diversity	BipedalWalker	GT QD Score6.09	6
Reinforcement Learning	BipedalWalker v3	Return273.2	6
Reinforcement Learning	bipedalwalker Sticky	AUC@T42,687,915.83	2
Reinforcement Learning	bipedalwalker Noisy	AUC@T32,301,685.83	2
Reinforcement Learning	bipedalwalker (Clean)	AUC@T9,392,950.11	2
Reinforcement Learning	BipedalWalker standard (test)	Length17	2
Interpretability Evaluation	BipedalWalker	Interpretability Score3.2	2

Showing 23 of 23 rows