ResearchBenchmarksContinuous Control on cartpoleBalanceFollow0.951Average RewardDDPG-0.52476-0.141630.24150.62463Sep 9, 2015Evaluation ResultsMethodMethodLinksAverage RewardBest RewardDDPGObservation Space=low-...Observation Space=low-dimensional2015.090.9511DDPGObservation Space=pixelsObservation Space=pixels2015.090.3350.996DPGSetup=original with re...Setup=original with replay buffer and batch normalization2015.09-0.4680.528