List your company

Continuous Control on MuJoCo HalfCheetah

13,144Average RewardTOP-TD3
-819.042,805.986,43110,056.02Feb 7, 2021Dec 18, 2021Oct 29, 2022Sep 9, 2023Jul 19, 2024May 30, 2025Apr 10, 2026

Evaluation results

MethodLinks
TOP-TD3Training steps=1M, Number of trials=102021.0213,144
ND TOP-TD3Training steps=1M, Number of trials=102021.0212,477
OACTraining steps=1M, Number of trials=102021.0211,723
QAvatarTraining steps=500k steps2026.0311,586
QR-TD3Training steps=1M, Number of trials=102021.0211,170
SACTraining steps=500k steps2026.0310,986
SACTraining steps=1M, Number of trials=102021.0210,815
FTTraining steps=500k steps2026.0310,756.8
TD3Training steps=1M, Number of trials=102021.029,566
CAT-SACTraining steps=500k steps2026.038,756.5
PARTraining steps=500k steps2026.038,097.4
IQ-LearnNumber of expert trajectories=102021.065,154
ExpertNumber of expert trajectories=102021.065,098
AGD-MBRLGuide=EAG2026.044,864.8
ValueDICENumber of expert trajectories=102021.064,840
AGD-MBRLGuide=SAG2026.044,642.25
GAILNumber of expert trajectories=102021.064,280
PolyGRADGuide=Diffuser Guide2026.044,188.43
PolyGRADGuide=None2026.043,879.16
BCNumber of expert trajectories=102021.062,701
PPO2026.042,408.2
TRPO2026.042,014.91
CATTraining steps=500k steps2026.0346.1
CMDTraining steps=500k steps2026.03-253.1
RandomNumber of expert trajectories=102021.06-282