ResearchBenchmarksReinforcement Learning on LunarLander v3 (Training Episodes to Threshold)Reinforcement Learning on LunarLander v3 (Training Episodes to Threshold)Follow290Episodes to Threshold (Score 200)SAC270.4402.7535667.3Mar 12, 2026Evaluation resultsMethodMethodLinksEpisodes to Threshold (Score 200)SACConfiguration=H-EARSConfiguration=H-EARS2026.03290TD3Configuration=H-EARSConfiguration=H-EARS2026.03350TD3Configuration=VanillaConfiguration=Vanilla2026.03540PPOConfiguration=H-EARSConfiguration=H-EARS2026.03560DDPGConfiguration=H-EARSConfiguration=H-EARS2026.03570SACConfiguration=VanillaConfiguration=Vanilla2026.03620DDPGConfiguration=VanillaConfiguration=Vanilla2026.03690PPOConfiguration=VanillaConfiguration=Vanilla2026.03780