List your company

Continuous Control on cartpole ParallelDouble

0.549Average RewardDDPG
0.173560.271030.36850.46597Sep 9, 2015

Evaluation results

MethodLinks
DDPGObservation Space=low-dimensional2015.090.5490.9
DPGSetup=original with replay buffer and batch normalization2015.090.1970.572
DDPGObservation Space=pixels2015.090.1880.323