Well-posedness, approximation and applications of reflected McKean-Vlasov SDEs (Michael Tretyakov, Nottingham)

13.01.2026 14:00

We establish well-posedness of reflected McKean-Vlasov SDEs and their particle approximations in smooth non-convex domains. We prove convergence of the interacting particle system to the corresponding mean-field limit with the optimal rate of convergence. We motivate this study by applications to sampling via reflected mean-field Langevin SDEs and optimization via reflected consensus-based optimization (CBO) models. We utilize reflection coupling to study long-time behaviour of reflected mean-field SDEs and investigate convergence of the reflected CBO models to the global minimum of a constrained optimization problem. Results of several numerical tests for CBO models are presented. The talk is based on a joint work with Piers Hinds and Akash Sharma.

Lieu

Conseil Général 7-9, Room 1-05, Séminaire d'analyse numérique

entrée libre