English

Transportation on spheres via an entropy formula

Probability 2024-09-24 v1

Abstract

The paper proves transportation inequalities for probability measures on spheres for the Wasserstein metrics with respect to cost functions that are powers of the geodesic distance. Let μ\mu be a probability measure on the sphere Sn{\bf S}^n of the form dμ=eU(x)dxd\mu =e^{-U(x)}dx where dxdx is the rotation invariant probability measure, and (n1)I+HessUκUI(n-1)I+{\hbox{Hess}}\,U\geq {\kappa_U}I, where κU>0\kappa_U>0. Then any probability measure ν\nu of finite relative entropy with respect to μ\mu satisfies Ent(νμ)(κU/2)W2(ν,μ)2{\hbox{Ent}}(\nu\mid\mu) \geq (\kappa_U/2)W_2(\nu, \mu )^2. The proof uses an explicit formula for the relative entropy which is also valid on connected and compact CC^\infty smooth Riemannian manifolds without boundary. A variation of this entropy formula gives the Lichn\'erowicz integral.

Keywords

Cite

@article{arxiv.2207.06191,
  title  = {Transportation on spheres via an entropy formula},
  author = {Gordon Blower},
  journal= {arXiv preprint arXiv:2207.06191},
  year   = {2024}
}

Comments

12 pages