대학원 인공지능학과 & AI융합네트워크학과에서는 Artificial Intelligence & AI Convergence Network Colloquium을 

6월12일(월) 오후1시 30분에 개최하고자 하오니 많은 참여 부탁드립니다.

 When : 2023년 6월 12일(월) 오후 1시 30분

 Where : 팔달관 407호 

 Speaker : 유재준 교수(UNIST AI 대학원 전임 & 전기전자공학)

 Title : Can We Find Strong Lottery Tickets in Generative Models?

 Abstract : Yes. In this talk, I show that it is possible to find a strong generative model without training, a.k.a the existence of strong lottery tickets, the subnetworks that achieve good performance without any weight update. Neural network pruning is considered the main cornerstone of model compression for reducing the costs of computation and memory. Unfortunately, pruning a generative model has not been extensively explored, and all existing pruning algorithms suffer from excessive weight-training costs, performance degradation, limited generalizability, or complicated training. To address these problems, we propose to find a strong lottery ticket via moment-matching scores. Our experimental results show that the discovered subnetwork can perform similarly or better than the trained dense model even when only 10% of the weights remain. To the best of our knowledge, we are the first to show the existence of strong lottery tickets in generative models and provide an algorithm to find it stably.

 Bio : 2021.07 - 현재       UNIST AI 대학원 전임 & 전기전자공학 겸임 조교수

           2019.12 - 2021.07  EPFL Postdoctoral researcher

           2018.02 - 2019.12  NAVER Search&Clova AI research scientist

           2018.02               KAIST 박사 바이오및뇌공학과 / 지도교수 예종철

 Host : 소프트웨어학과 손경아 교수(kasohn@ajou.ac.kr)