1. 일시: 2023년 9월 21일(목), 14:00-16:00

2. 장소: 판교 테크노밸리 산업수학혁신센터 세미나실

3. 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소 [무료주차는 2시간 지원됩니다.]

4. 발표자: 권도현 교수(서울시립대학교)

5. 주요내용: Applications of Optimal Transport to Generative Models

Over the past few decades, optimal transport theory has gained increasing interest across multiple fields, including partial differential equations, probability, and machine learning. In this talk, we explore the diverse applications of optimal transport theory within various machine learning problems, with a specific focus on generative models. Our discussion begins by examining gradient flows in the space of probability measures equipped with the distance arising from the Monge-Kantorovich optimal transport problem. We then analyze a score-based generative model based on the Fokker-Planck equations that underlie both the forward and reverse processes of the model. Additional discussions about De Giorgi's minimizing ments and Wasserstein dictionary learning will be provided.

*유튜브 스트리밍 예정입니다.