ResearchBenchmarksInvestment Policy Equilibrium Recovery on Merton Problem Case 2Follow0MAE (L1)PG-DPO-0.021560.123970.26950.41503May 20, 2026Evaluation ResultsMethodMethodLinksMAE (L1)Max Error (Linf)PG-DPO2026.0500BSVIE2026.050.04080.244PINN2026.050.04730.305DPO2026.050.3121.11PPO2026.050.5392.22