List your company
ResearchFormerly WizWand

Reinforcement Learning on Hopper (Average Episode Reward)

2,743.9Avg Episode RewardTD3
-60.512667.55651,395.6252,123.6935Nov 2, 2023Mar 17, 2024Jul 31, 2024Dec 14, 2024Apr 29, 2025Sep 12, 2025Jan 26, 2026
Updated 8mo ago

Evaluation results

MethodLinks
TD32023.112,743.9
SAC2023.112,613.16
TRPO2023.112,593.56
PPO2023.112,586.56
ACKTR2023.112,583.88
ESPL2023.112,442.48
DSP2023.112,122.4
A2C2023.112,104.98
DDPG2023.111,678.84
SMACbatch size=1000, seeds=52026.01331.2
AC-KFACbatch size=1000, seeds=52026.01305
AC-Adambatch size=1000, seeds=52026.01303.8
AC-SGDbatch size=1000, seeds=52026.01180.5
AC-CGbatch size=1000, seeds=52026.01172.7
Regression2023.1147.35