학술행사
- 발표자 김연응 교수 (가천대학교)
- 개최일시 2022-04-01 10:30
- 장소 국가수리과학연구소 산업수학혁신센터(광교)
* 사전 신청 부탁드립니다.(https://icim.nims.re.kr/post/event/910)
1. 일시 : 2022년 4월 1일(금), 10:30-12:30
2. 장소 : 광교 테크노밸리 산업수학혁신센터 세미나실
3. 발표자 : 김연응 교수 (가천대학교)
4. 주요내용 :Some Mathematical Aspects of Machine Learning
Various subjects in the field of machine learning require a deep understanding of rigorous mathematics. In this talk I will introduce some fields in machine learning together with particular problems I have been tackling recently. In the first part the notion of Wasserstein Generative Adversarial Networks which has received great deals of attention is introduced. Recently we developed a new algorithm called CoWGAN, which is not only more efficient but also mathematically plausible in the eyes of the theory of optimal transport. In the second part of the talk, which is unrelated to the first part, I will give a brief introduction to multi-armed bandit problems and present a new result on variance-aware problems.
5. 유튜브 실시간 스트리밍 : 현장 참석이 어려운 분들을 위해 온라인으로 실시간 방송할 예정입니다. 주소는 당일 사전신청 페이지에 업데이트 하겠습니다.
* 사전 신청 부탁드립니다.(https://icim.nims.re.kr/post/event/910)
1. 일시 : 2022년 4월 1일(금), 10:30-12:30
2. 장소 : 광교 테크노밸리 산업수학혁신센터 세미나실
3. 발표자 : 김연응 교수 (가천대학교)
4. 주요내용 :Some Mathematical Aspects of Machine Learning
Various subjects in the field of machine learning require a deep understanding of rigorous mathematics. In this talk I will introduce some fields in machine learning together with particular problems I have been tackling recently. In the first part the notion of Wasserstein Generative Adversarial Networks which has received great deals of attention is introduced. Recently we developed a new algorithm called CoWGAN, which is not only more efficient but also mathematically plausible in the eyes of the theory of optimal transport. In the second part of the talk, which is unrelated to the first part, I will give a brief introduction to multi-armed bandit problems and present a new result on variance-aware problems.
