List your company

Locomotion Control on D4RL Walker2d medium-expert

111.2Normalized ReturnRepb-SDE
-4.7625.34555.4585.555Feb 16, 2021Sep 29, 2021May 13, 2022Dec 24, 2022Aug 7, 2023Mar 19, 2024Oct 31, 2024

Evaluation results

MethodLinks
Repb-SDEalgorithm class=model-based2021.10111.2
ROMI-CQLbase algorithm=CQL2021.10109.7
CQLalgorithm class=model-free2021.10109.5
Fisher-BRCimplementation=author-provided (re-run)2021.06103.6
COMBO2021.02103.3
TD3+BCalpha=2.52021.06101.1
KIOAmount of data used=100k, Evaluation=Average score over 100 episodes2024.10100.1
CQL2021.0298.7
MOREL2021.0295.6
TeacherEvaluation=Average score over 100 episodes2024.1082.7
BRAC-v2021.0281.6
BRAC-p2021.0276.9
BC2021.1073.4
CQLimplementation=author-provided (re-run)2021.0668.1
MOPO2021.0244.6
AWACimplementation=reported2021.0642.7
BEAR2021.0240.1
BC(TD3+BC)Evaluation=Average score over 100 episodes2024.1012
BCimplementation=own2021.0612
BC(CQL)Evaluation=Average score over 100 episodes2024.1011.3
BC2021.026.4
SAC-off2021.02-0.1
BRAC-pimplementation=reported2021.06-0.3