Trajectory Inference via Multi-marginal Schrödinger Bridges
이재문
27동 325호
0
1257
05.11 22:07
| 구분 | HYKE,기타 |
|---|---|
| 일정 | 2026-06-01(월) 15:30~17:30 |
| 세미나실 | 27동 325호 |
| 강연자 | 김영헌 (Univ. of British Columbia) |
| 담당교수 | 하승열 |
| 기타 | HYKE-Hwarang 세미나 |
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%
% HYKE-Hwarang 세미나
%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
일시: 2026년 6월 1일 (월) 16:00 - 17:00
장소: 27동 325
연사: 김영헌 (Univ. of British Columbia)
발표제목: Trajectory Inference via Multi-marginal Schrödinger Bridges
초록: Trajectory inference arises in important scientific problems. In particular, biological development can be interpreted as a curve in the space of gene-expression distributions, and the goal is to infer this trajectory from observed data. There has been progress by using optimal transport (OT) as a way to interpolate between distributions. More recently, Schrödinger bridges, a stochastic generalization of OT, have been considered. In this talk, we discuss stability of such OT-based methods. This is joint work with Geoffrey Schiebinger and Rentian Yao.
