Computational optimal transport (OT) has recently emerged as a powerful framework with applications in various fields. In this paper we focus on a relaxation of the original OT problem, the entropic OT problem, which allows to implement efficient and practical algorithmic solutions, even in high dimensional settings. This formulation, also known as the Schr\"odinger Bridge problem, notably connects with Stochastic Optimal Control (SOC) and can be solved with the popular Sinkhorn algorithm. In the case of discrete-state spaces, this algorithm is known to have exponential convergence; however, achieving a similar rate of convergence in a more general setting is still an active area of research. In this work, we analyze the convergence of the Sinkhorn algorithm for probability measures defined on the $d$-dimensional torus $\mathbb{T}_L^d$, that admit densities with respect to the Haar measure of $\mathbb{T}_L^d$. In particular, we prove pointwise exponential convergence of Sinkhorn iterates and their gradient. Our proof relies on the connection between these iterates and the evolution along the Hamilton-Jacobi-Bellman equations of value functions obtained from SOC-problems. Our approach is novel in that it is purely probabilistic and relies on coupling by reflection techniques for controlled diffusions on the torus.
翻译:计算最优输运(OT)最近在各个领域展现出强大的应用。本文重点研究原始 OT 问题的一种松弛形式,也就是熵正则化 OT 问题,这种形式使得即便在高维背景下,实现高效实用的算法解决方案成为了可能,被称为Schr"odinger Bridge问题,尤其与随机最优控制(SOC)相联系,可以用流行的Sinkhorn算法解决。在离散状态空间中,该算法被认为具有指数收敛;然而,在更通用的情况下实现类似的收敛速率仍然是一个活跃的研究领域。在本文中,我们分析了Sinkhorn算法在定义在Toru L^d上且相对Haar度量具有密度的概率测度上的收敛性。特别地,我们证明了Sinkhorn迭代及其梯度的点位指数收敛。我们的证明依赖于这些迭代与由SOC问题得到的价值函数的 Hamilton-Jacobi-Bellman 方程沿着其演化的联系。我们的方法是新颖的,纯粹基于概率,并依赖於在Toru上受控扩散的反射耦合技术。