Id: arxiv/2608.07433 Url: https://arxiv.org/abs/2608.07433 Tags: math.OC, cs.LG Score: 0 Title: Wasserstein Policy Gradient for Entropy-Regularized Linear-Quadratic Control Domain: math.OC Source: arxiv Authors: Zhaoyu Zhu, Rui Gao, Shuang Li Reviews: 0 Subject: math.OC Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-08-07 Page: https://delphi.science/r/arxiv/2608.07433 Wasserstein policy gradient (WPG) updates state-conditional action laws by transport in the action space. We study entropy-regularized discounted linear-quadratic (LQ) control. A Bellman verification argument shows that the unrestricted problem has a linear-Gaussian optimal policy, and the discounted-occupancy-weighted statewise Wasserstein gradient is tangent to this policy class. WPG therefore reduces exactly to a finite-dimensional ODE for the feedback gain and action covariance. We prove that this ODE is globally well posed and converges exponentially from every admissible…