ResearchBenchmarksContinuous Control on MuJoCo InvertedDoublePendulum 10M framesFollow3,684.92Mean ScorePPO2,516.8442,820.09453,123.3453,426.5955Jul 2, 2018Evaluation ResultsMethodMethodLinksMean ScorePPOTotal frames=10M, Numb...Total frames=10M, Number of trials=3, Episode window=All episodes2018.073,684.92POP3DTotal frames=10M, Numb...Total frames=10M, Number of trials=3, Episode window=All episodes2018.072,561.77