List your company

Reinforcement Learning on Gymnasium Humanoid

6,367ReturnCAPO
191.481,794.743,3985,001.26Mar 13, 2026

Evaluation results

MethodLinks
CAPOBackbone=(64, 64) MLP, Training steps=4M, Seeds=8, Compute-matched=true2026.036,367
TRPOBackbone=(64, 64) MLP, Training steps=4M, Seeds=8, Compute-matched=false2026.033,730
PPO-SWABackbone=(64, 64) MLP, Training steps=4M, Seeds=8, Compute-matched=false2026.03750
PPOBackbone=(64, 64) MLP, Training steps=4M, Seeds=8, Compute-matched=false2026.03739
Best-of-KBackbone=(64, 64) MLP, Training steps=4M, Seeds=8, Compute-matched=true2026.03716
CAPO-AvgBackbone=(64, 64) MLP, Training steps=4M, Seeds=8, Compute-matched=true2026.03696
PPO-K×Backbone=(64, 64) MLP, Training steps=4M, Seeds=8, Compute-matched=true2026.03429