ResearchBenchmarksReinforcement Learning on Hopper v5 (Coefficient of Variation)Reinforcement Learning on Hopper v5 (Coefficient of Variation)Follow10.6Coefficient of VariationSAC10.2212.78515.3517.915Mar 12, 2026Evaluation resultsMethodMethodLinksCoefficient of VariationSACVariant=H-EARSVariant=H-EARS2026.0310.6TD3Variant=H-EARSVariant=H-EARS2026.0311.3TD3Variant=VanillaVariant=Vanilla2026.0312.8PPOVariant=H-EARSVariant=H-EARS2026.0312.9PPOVariant=VanillaVariant=Vanilla2026.0315.7SACVariant=VanillaVariant=Vanilla2026.0316.1DDPGVariant=H-EARSVariant=H-EARS2026.0319.6DDPGVariant=VanillaVariant=Vanilla2026.0320.1