|
 코드번호 : 35 |
|
| 발표자 : 유재준 |
|
| 소속 : UNIST |
|
| 부서 : |
|
| 직위 : 교수 |
|
| 세션시간 : 14:00~15:50 |
|
| 발표자약력 : |
2007-2011 한국과학기술원 학사 졸업
2011-2013 한국과학기술원 석사 졸업 (지도교수: 예종철)
2013-2018 한국과학기술원 박사 졸업 (지도교수: 예종철)
2018-2019 네이버 Clova AI Lab Research Scientist
2020-2021 EPFL Postdoctoral Researcher
2021.07-현재 UNIST 인공지능대학원 전임 및 전기전자공학과 겸임 |
|
|
| 강연요약 : |
We find ourselves in an era dominated by large generative models, which, while demonstrating remarkable performance, demand vast datasets and substantial computational resources. However, these resources remain inaccessible to the majority of researchers and all but a few well-resourced companies. As we navigate this landscape, the quest for scalable, efficient, and privacy-preserving mechanisms becomes more crucial than ever. In this seminar, I will delve into our recent research efforts, making generative models more accessible, efficient, and compliant with the stringent requirements of practical applications. |
|
|
|
