AI RESEARCH
Well-Posed KL-Regularized Control via Wasserstein and Kalman-Wasserstein KL Divergences
arXiv CS.LG
•
ArXi:2602.02250v2 Announce Type: replace-cross Kullback-Leibler (KL) divergence regularization is widely used in reinforcement learning, but it becomes infinite under mismatch and can degenerate in low-noise regimes. Using a unified information-geometric framework, we