Statistical manifolds from optimal transport
نویسنده
چکیده
Divergences, also known as contrast functions, are distance-like quantities defined on manifolds of non-negative or probability measures and they arise in various theoretical and applied problems. Using ideas in optimal transport, we introduce and study a parameterized family of $L^{(\pm \alpha)}$-divergences which includes the Bregman divergence corresponding to the Euclidean quadratic cost, and the $L$-divergence introduced by Pal and Wong in connection with portfolio theory and a logarithmic cost function. Using this unified framework which elucidates the arguments in our previous work, we prove that these divergences induce geometric structures that are dually projectively flat with constant curvatures, and the generalized Pythagorean theorem holds true. Conversely, we show that if a statistical manifold is dually projectively flat with constant curvature $\pm \alpha$ with $\alpha>0$, then it is locally induced by an $L^{(\mp \alpha)}$-divergence. We define in this context a canonical divergence which extends the one for dually flat manifolds. Finally, we study generalizations of exponential family and show that the $L^{(\pm \alpha)}$-divergence of the corresponding potential functions gives the R\'{e}nyi divergence.
منابع مشابه
Regularity of optimal transport maps on multiple products of spheres
This article addresses regularity of optimal transport maps for cost=“squared distance” on Riemannian manifolds that are products of arbitrarily many round spheres with arbitrary sizes and dimensions. Such manifolds are known to be non-negatively cross-curved [KM2]. Under boundedness and non-vanishing assumptions on the transfered source and target densities we show that optimal maps stay away ...
متن کاملIsoperimetric-type inequalities on constant curvature manifolds
By exploiting optimal transport theory on Riemannian manifolds and adapting Gromov’s proof of the isoperimetric inequality in the Euclidean space, we prove an isoperimetric-type inequality on simply connected constant curvature manifolds.
متن کاملNecessary and sufficient conditions for continuity of optimal transport maps on Riemannian manifolds
In this paper we investigate the regularity of optimal transport maps for the squared distance cost on Riemannian manifolds. First of all, we provide some general necessary and sufficient conditions for a Riemannian manifold to satisfy the so-called Transport Continuity Property. Then, we show that on surfaces these conditions coincide. Finally, we give some regularity results on transport maps...
متن کاملOptimal transportation on non-compact manifolds
In this work, we show how to obtain for non-compact manifolds the results that have already been done for Monge Transport Problem for costs coming from Tonelli Lagrangians on compact manifolds. In particular, the already known results for a cost of the type dr , r > 1, where d is the Riemannian distance of a complete Riemannian manifold, hold without any curvature restriction.
متن کاملStatistical cosymplectic manifolds and their submanifolds
In this paper, we introduce statistical cosymplectic manifolds and investigate some properties of their tensors. We define invariant and anti-invariant submanifolds and study invariant submanifolds with normal and tangent structure vector fields. We prove that an invariant submanifold of a statistical cosymplectic manifold with tangent structure vector field is a cosymplectic and minimal...
متن کامل