학술행사
- 발표자 NO ALBERT(연세대학교)
- 개최일시 2025-01-16 10:30-12:30
- 장소 국가수리과학연구소 산업수학혁신센터(판교)
유튜브 스트리밍 예정입니다.
-
일시: 2025년 1월 16일(목), 10:30-12:30
-
장소: 판교 테크노밸리 산업수학혁신센터 세미나실
-
경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소 / 무료주차는 2시간 지원됩니다.
-
발표자: NO ALBERT(연세대학교)
-
주제: Understanding Memorization in Generative Models via Sharpness in Probability Landscapes
In this talk, we introduce a geometric framework to analyze memorization in diffusion models using the eigenvalues of the Hessian of the log probability density. We propose that memorization arises from isolated points in the learned probability distribution, characterized by sharpness in the probability landscape, as indicated by large negative eigenvalues of the Hessian. Through experiments on various datasets, we demonstrate that these eigenvalues effectively detect and quantify memorization. Our approach provides a clear understanding of memorization in diffusion models and lays the groundwork for developing strategies to ensure secure and reliable generative models
유튜브 스트리밍 예정입니다.
-
일시: 2025년 1월 16일(목), 10:30-12:30
-
장소: 판교 테크노밸리 산업수학혁신센터 세미나실
-
경기 성남시 수정구 대왕판교로 815, 기업지원허브 231호 국가수리과학연구소 / 무료주차는 2시간 지원됩니다.
-
발표자: NO ALBERT(연세대학교)
-
주제: Understanding Memorization in Generative Models via Sharpness in Probability Landscapes
In this talk, we introduce a geometric framework to analyze memorization in diffusion models using the eigenvalues of the Hessian of the log probability density. We propose that memorization arises from isolated points in the learned probability distribution, characterized by sharpness in the probability landscape, as indicated by large negative eigenvalues of the Hessian. Through experiments on various datasets, we demonstrate that these eigenvalues effectively detect and quantify memorization. Our approach provides a clear understanding of memorization in diffusion models and lays the groundwork for developing strategies to ensure secure and reliable generative models
