ResearchBenchmarksContinuous Control on fixedReacherSingleFollow0.954Average RewardDDPG0.606640.696820.7870.87718Sep 9, 2015Evaluation ResultsMethodMethodLinksAverage RewardBest RewardDDPGObservation Space=low-...Observation Space=low-dimensional2015.090.9541DDPGObservation Space=pixelsObservation Space=pixels2015.090.8270.995DPGSetup=original with re...Setup=original with replay buffer and batch normalization2015.090.620.999