<University of Central Florida 강수연 교수님 초청 강연 안내>

1. 연사: University of Central Florida 강수연 교수님

2. 주제: Low-rank regularization for finite mixtures of multivariate response regression models

3. 날짜: 2026년 6월 5일(금) 오후 4시

4. 장소: 사이언스홀(자연대 1호관 100호)

5. 발표 초록 :

Multivariate regression models with multiple continuous outcomes arise naturally in many scientific applications. In practice, populations may exhibit heterogeneous subgroup structure, while associations among multiple responses are often driven by a smaller number of latent components. Motivated by these settings, this talk considers reduced-rank estimation in finite mixtures of multivariate response regression models.

The proposed framework combines finite mixture modeling with low-rank regularization to simultaneously account for subgroup heterogeneity and low-dimensional dependence among multiple responses. I will discuss computationally efficient EM-type estimation algorithms that enable subgroup identification, parameter estimation, and rank selection within a unified framework, along with several key theoretical properties of the proposed estimators. The proposed methods are illustrated through simulation studies and a real data application.