明平兵:Barron Type Spaces in Neural Network Approximation
Academy of Mathematics and Systems Science, CAS Colloquia & Seminars
Speaker:
明平兵,中国科学院数学与系统科学研究院
Inviter:
中科院数学院学术报告会组委
Title:
Barron Type Spaces in Neural Network Approximation
Language:
Chinese
Time & Venue:
2023.03.31 10:40-12:00 腾讯会议:877-4221-8481
Abstract:
We shall discuss various Barron type spaces arising from neural networks. The relations among these spaces will be clarified, and we shall also establish the relationship between Barron type spaces and the classical function spaces such as Besov spaces, Sobolev spaces and Bessel potential spaces. As an application, certain new approximation rates for the shallow and deep neural networks with the Barron class as the target function space will be proved. This is a joint work with Yulei Liao (AMSS, CAS) and Yan Meng (RUC).