커뮤니티
  • HOME
  • 커뮤니티
  • 세미나
세미나
[2024.04.29(월)] Artificial Intelligence & AI Convergence Network Colloquium 개최 안내
  • 글쓴이 관리자
  • 작성일 2024-04-23 16:45:51
  • 조회수 127

Artificial Intelligence & AI Convergence Network 사업단에서 공동으로 

개최하는 Colloquium을 4월 29일(월) 오후 3시에 개최하오니 많은 참여 부탁드립니다.


▶️  When : 2024년 4월 29일(월) 오후 3시

▶️  Where : 팔달관 407호 
▶️  Speaker : 강병곤 교수(SUNY Korea)
▶️  Title : On the efficiency of pre-trained word embeddings on Transformers

▶️  Abstract : In this talk, I will discuss a counter-intuitive phenomenon where state-of-the-art pretrained word embedding vectors perform poorly compared to randomly initialized vectors.
More specifically, such a tendency is observed almost exclusively in Transformer architectures, making it a problem to most modern NLP systems that are heavily dependent on Transformers. 
I also describe a very simple remedy that somewhat alleviates this shortcoming, as well as numerous empirical results to back this claim.
If time permits, I will also share a failure case of another popular deep learning system (autograd), leading us to the discussion of whether we should take prevalent technologies for granted in this field.

▶️  Bio : Byungkon Kang got his PhD at KAIST in 2013 before joining Samsung Advanced Institute of Technology. After that, he spent a few years at Ajou University as a research professor.He is currently an assistant professor at SUNY Korea, working on various AI/ML topics. 

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

콜로키움 포스터.png
목록





이전글 [2024.04.18(목)] AI Convergence Network & Artificial Intelligence Colloquium ...
다음글 [2024.05.2(목)] Artificial Intelligence & AI Convergence Network Colloquium 개...