A Proposal on Machine Learning via Dynamical Systems

被引：387
作者：
Weinan, E. [1 ,2 ,3 ,4 ,5 ]
机构：
[1] BIBDR, Beijing, Peoples R China
[2] Princeton Univ, Dept Math, Princeton, NJ 08544 USA
[3] Princeton Univ, PACM, Princeton, NJ 08544 USA
[4] Peking Univ, Ctr Data Sci, Beijing, Peoples R China
[5] Peking Univ, BICMR, Beijing, Peoples R China
来源：
COMMUNICATIONS IN MATHEMATICS AND STATISTICS | 2017年 / 5卷 / 01期
关键词：
Deep learning; Machine learning; Dynamical systems;
D O I：
10.1007/s40304-017-0103-z
中图分类号：
O1 [数学];
学科分类号：
0701 ; 070101 ;
摘要：
We discuss the idea of using continuous dynamical systems to model general high-dimensional nonlinear functions used in machine learning. We also discuss the connection with deep learning.
引用
收藏
页码：1 / 11
页数：11
相关论文
共 14 条
[1]  
Almeida L. B., 1987, P ICNN 87 SAN DIEG I
[2]  
[Anonymous], 1987, P IEEE C NEUR INF PR
[3]  
Fan J., 1996, LOCAL POLYNOMIAL MOD
[4]  
HAN J, UNPUB
[5]  
Hastie T, 2013, ELEMENTS STAT LEARNI
[6]  
He K., 2016, PROC CVPR IEEE, P630, DOI [10.1007/978-3-319-46493-0_38, DOI 10.1007/978-3-319-46493-0_38, DOI 10.1109/CVPR.2016.90]
[7]  
Lambert J.D., 1992, Numerical Methods for Ordinary Differential Systems: The Initial Value Problem
[8]  
LeCun Y., 1989, P 1988 CONN MOD SUMM
[9]  
LI QS, UNPUB
[10]  
Ming P., CALCULUS VARIATIONS
12