← Back to all news
arXiv Machine Learning By Viktor Stein, Adwait Datar, Nihat Ay

Well-Posed KL-Regularized Control via Wasserstein and Kalman-Wasserstein KL Divergences

Read the original on arXiv Machine Learning →

arXiv:2602. 02250v2 Announce Type: replace-cross Abstract: Kullback-Leibler (KL) divergence regularization is widely used in reinforcement learning, but it becomes infinite under support mismatch and can degenerate in low-noise regimes.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Jul 27

Trajectory-Regularized Stochastic Optimal Control via KL Divergence

arXiv:2607. 22201v1 Announce Type: cross Abstract: We introduce trajectory-regularized stochastic optimal control (TRSOC), which augments standard stochastic optimal control (SOC) with a Kullback--Leibler (KL) divergence between controlled and reference trajectory distributions.

By Mintae Kim, Koushil Sreenath
More like this →
arXiv Machine Learning
Sep 17

Preservation of Log-Concavity and Convergence of Wasserstein-Fisher-Rao Gradient Flows

The paper investigates Wasserstein-Fisher-Rao (WFR) gradient flows for sampling from probability distributions known only up to a normalisation constant. It demonstrates that for strongly log-concave targets satisfying certain curvature conditions, WFR flows preserve strong log-concavity—unlike pure Wasserstein flows, which only do so in the Gaussian case. Leveraging this property, the authors derive explicit non-asymptotic convergence rates for the symmetrised Kullback-Leibler divergence, showing an additive decomposition into Wasserstein and Fisher‑Rao contributions and eliminating the need for a warm start.

By Francesca Romana Crucinio, Sahani Pathiraja
More like this →
arXiv Machine Learning
Sep 11

Fisher-Rao Gradient Flows of Linear Programs and State-Action Natural Policy Gradients

The paper investigates a natural gradient method based on the Fisher information matrix of state-action distributions, which follows a Fisher‑Rao gradient flow within the state-action polytope under a linear potential. It establishes linear convergence rates for Fisher‑Rao gradient flows of linear programs, with the rate tied to the program’s geometry, and provides improved error bounds for entropic regularization. Additionally, the authors extend their analysis to perturbed flows, proving sublinear convergence for both perturbed Fisher‑Rao and natural gradient flows, thereby encompassing state‑action natural policy gradients.

By Johannes M\"uller, Semih \c{C}ayc{\i}, Guido Mont\'ufar
reinforcement-learning
More like this →