报告题目 Title：Learning with norm-based neural networks: model capacity and computational-statistical gaps

报告人 Speaker：刘方辉 博士

报告人所在单位 Affiliation：University of Warwick, UK

时间 Time：2024-09-04 14:30-15:30

地点 Venue：光华楼东主楼 2001

报告摘要 Abstract：

In this talk, I will discuss some fundamental questions in modern machine learning:

  -What is the suitable model capacity of over-parameterized models?

  -What is the suitable function space for feature learning?

  -Which function can be learned by two-layer neural networks, statistical and/or computational efficiently?

  -What is the computational-statistical gap behind this?

My talk will partly answer the above questions, both theoretically and empirically.

海报 Poster：刘方辉 博士 学术报告.JPG