期刊
NEURAL NETWORKS
卷 129, 期 -, 页码 1-6出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.neunet.2019.12.013
关键词
Deep ReLU networks; Curse of dimensionality; Approximation theory; Kolmogorov-Arnold superposition theorem
资金
- Department of Mathematics at the National University of Singapore
- Ministry of Education in Singapore [MOE2018-T2-2-147]
We prove a theorem concerning the approximation of multivariate functions by deep ReLU networks, for which the curse of the dimensionality is lessened. Our theorem is based on a constructive proof of the Kolmogorov-Arnold superposition theorem, and on a subset of multivariate continuous functions whose outer superposition functions can be efficiently approximated by deep ReLU networks. (C) 2019 Published by Elsevier Ltd.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据