{"id":"arxiv/2610.03264","url":"https://arxiv.org/abs/2610.03264","tags":["math.OC","eess.SY"],"score":0,"title":"Wasserstein Contraction of Stochastic Systems on Manifolds: A Differential Approach","domain":"math.OC","source":"arxiv","authors":["Dongjun Wu"],"reviews":0,"subject":"math.OC","summary":"Contraction analysis of dynamical systems turns a global convergence problem into a local one. If the variational system is stable in a uniform sense, then all trajectories of the original system converge toward each other. This paper develops a stochastic analogue for continuous-time stochastic systems, where convergence is measured by Wasserstein distance between probability laws. The key step is to employ mean-square differentiation to construct stochastic variational dynamics. Its uniform $L_2$ stability can then be integrated along curves of initial conditions, yielding Wasserstein…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-10-02","reactions":{}}