일시: 2023년 3월 6일(월), 14:00-16:00

장소: 판교 테크노밸리 산업수학혁신센터 세미나실

• 경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소
• 무료주차는 2시간 지원됩니다.

발표자: 노영균 교수(한양대학교)

주요내용: Techniques for f-divergence Estimation and their Applications to Machine Learning

In this talk, various f-divergence estimation techniques and their applications will be explained based on the connection to loss functions. It is known that loss functions are connected to their corresponding f-divergences which measure the theoretical limit of the associated loss functions. In this regard, researchers have focused on the problems that tackle the optimization of f-divergences instead of loss functions. I will clarify the problems that we can benefit from optimizing the f-divergences and explain how novel techniques for f-divergence estimation can be used for real applications.