报告人：杨士源 博士
报告题目：Asymptotics of Multi-Scale McKean--Vlasov Diffusions with Super-Linear Kernels
报告时间：2026年10月11日（周日）上午9：00
报告地点：云龙校区6号楼304报告厅
主办单位：数学与统计学院、数学研究院、科学技术研究院
报告人简介：
杨士源，天津大学在读博士生（二年级），师从王凤雨教授，已在SIAM J. Control Optim.、Discrete Contin. Dyn. Syst. Ser. S杂志上发表论文。
报告摘要：
We study multi-scale McKean--Vlasov systems with locally Lipschitz coefficients and super-linear kernels, where the interactions depend on the laws of both the slow component and the fast oscillating process. By combining a lifted semigroup approach with a generalized Khasminskii discretization, we establish a functional law of large numbers with an explicit convergence rate. We also prove a large deviation principle with an explicitly characterized rate function by introducing lifted viable pairs and employing a feedback-control argument. The results apply to multi-scale consensus-based optimization and ensemble Kalman sampler models arising in machine learning and optimization.