Seminar on Statistical Inference in Non-Euclidean spaces

Abstract

This seminar discusses the statistical methods in Non-Euclidean spaces. First, we focus on basic theoretical tools and statistical models in Euclidean space, including U-statistics, concentration inequalities, limit theories of weakly dependent sequences, graphical models, empirical processes and theoretical frameworks of neural networks. Then, we focus on the generalization of classical statistical methods to non-Euclidean space, aiming to propose a statistical model that is theoretically guaranteed, efficient, and widely applicable. We further discuss two sample testing, change point detection, PCA, graphical models, generative models in non-Euclidean space.

Time and Location

Time: 2:00pm-4:00pm, Thursday, 2025 Spring

Location: Online, Tencent Meetings

Learning Materials

Materials are temporarily stored on Github: PPT.