ResearchBenchmarksMulti-Objective Reinforcement Learning on Lunar Lander 4dFollow1.24Hypervolume (HV)SPFT0.39760.61630.8351.0537Aug 4, 2025Sep 4, 2025Oct 5, 2025Nov 6, 2025Dec 7, 2025Jan 7, 2026Feb 8, 2026Evaluation ResultsMethodMethodLinksHypervolume (HV)Expected Utility (EU)Sparsity (SP)Compute Time (CT) (hours)Environment StepsSPFT2025.081.242.361.75—480,000D3PO2026.021.232.393210—C-MORL2026.021.122.3510420—C-MORL2025.081.122.351.04—500,000GPI-LS2026.021.061.81135—GPI-LS2025.081.061.690.13—500,000PCN2026.020.781.4437—Envelope2025.080.43-2.840.19—500,000