[finished] ipi seminar 10:30-12:00, Thursday Dec. 16, 2021

固定ページ 2021/11/22

知の物理学研究センター / Institute for Physics of Intelligence (ipi)

 

【Speaker】Sho YAIDA  (Meta AI)

【Date】10:30-12:00 JST, Thursday, Dec. 16

【Title】"Effective Theory of Deep Neural Networks"

【Abstract】
Large neural networks perform extremely well in practice, providing the backbone of modern machine learning. The goal of this talk is to provide a blueprint for theoretically analyzing these large models from first principles. In particular, we'll overview how the statistics and dynamics of deep neural networks drastically simplify at large width and become analytically tractable. In so doing, we'll see that the idealized infinite-width limit is too simple to capture several important aspects of deep learning such as representation learning. To address them, we'll step beyond the idealized limit and systematically incorporate finite-width corrections.

*To receive the Zoom invitation and monthly reminders, please register via this google form: https://forms.gle/dqxhpsZXLNYvbSB38
Your e-mail addresses will be used for this purpose only, you can unsubscribe anytime, and we will not send more than three e-mails per month.
 
 
Tilman HARTWIG, Takashi TAKAHASHI & Ken NAKANISHI
 
Related Links :
  • Bookmark