List your company

Reinforcement Learning on HalfCheetah v4

10,554Max ReturnReLU
1,611.043,932.776,254.58,576.23Jan 29, 2026Feb 22, 2026Mar 18, 2026Apr 11, 2026May 5, 2026May 29, 2026Jun 22, 2026

Evaluation results

MethodLinks
ReLUFrame=ANN2026.0110,554
C-DSACNumber of runs=100, Selection=Best models in training runs2026.0410,023
S-PLIFFrame=Vanilla2026.019,828
S-PLIFFrame=PT2026.019,405
PLIFFrame=Vanilla2026.019,252
PLIFFrame=PT2026.019,219
PDAEnvironment steps=1M, Number of seeds=10, Tests per epoch=102026.035,174.6
TRPOEnvironment steps=1M, Number of seeds=10, Tests per epoch=102026.034,496.8
PPOEnvironment steps=1M, Number of seeds=10, Tests per epoch=102026.034,067.5
NPGEnvironment steps=1M, Number of seeds=10, Tests per epoch=102026.033,556.4
Adaptive βNumber of seeds=5, Evaluation window=last 10% of training2026.062,207
Fixed βNumber of seeds=5, Evaluation window=last 10% of training2026.062,111
PPO-ClipNumber of seeds=5, Evaluation window=last 10% of training2026.061,955
per-sample PPO-KLNumber of seeds=5, Evaluation window=last 10% of training2026.061,955