ResearchBenchmarksContinuous Control on MuJoCo HalfCheetah Hard ConstraintFollow2,786.71ReturnPPOLag2,486.26442,564.26472,642.2652,720.2653Jun 28, 2024Evaluation ResultsMethodMethodLinksReturnConstraint Violation RatePPOLagCost Knowledge=KnownCost Knowledge=Known2024.062,786.710.42SIMKCCost Knowledge=KnownCost Knowledge=Known2024.062,497.820.06