DEEP RELU NETWORKS OVERCOME THE CURSE OF DIMENSIONALITY FOR GENERALIZED BANDLIMITED FUNCTIONS

被引：17

作者：

Montanelli, Hadrien ^{[1
]}

Yang, Haizhao ^{[2
]}

Du, Qiang ^{[3
]}

机构：

[1] Ecole Polytech, Ctr Math Appl, Palaiseau, France

[2] Purdue Univ, Dept Math, W Lafayette, IN 47907 USA

[3] Columbia Univ, Dept Appl Phys & Appl Math, New York, NY USA

来源：

JOURNAL OF COMPUTATIONAL MATHEMATICS | 2021年 / 39卷 / 06期

关键词：

Machine learning; Deep ReLU networks; Curse of dimensionality; Approxima-tion theory; Bandlimited functions; Chebyshev polynomials; ERROR-BOUNDS; OPTIMAL APPROXIMATION; SUPERPOSITION; SMOOTH;

D O I：

10.4208/jcm.2007-m2019-0239

中图分类号：

O29 [应用数学];

学科分类号：

070104 ;

摘要：

We prove a theorem concerning the approximation of generalized bandlimited multivariate functions by deep ReLU networks for which the curse of the dimensionality is overcome. Our theorem is based on a result by Maurey and on the ability of deep ReLU networks to approximate Chebyshev polynomials and analytic functions efficiently.

引用

页码：801 / 815

页数：15

共 50 条

[1] Deep ReLU neural networks overcome the curse of dimensionality for partial integrodifferential equations
Gonon, Lukas
Schwab, Christoph
ANALYSIS AND APPLICATIONS, 2023, 21 (01) : 1 - 47
[2] Functions bandlimited in frequency are free of the curse of dimensionality
Ignacio Mulero-Martinez, Juan
NEUROCOMPUTING, 2007, 70 (7-9) : 1439 - 1452
[3] Optimal Bump Functions for Shallow ReLU networks: Weight Decay, Depth Separation, Curse of Dimensionality
Wojtowytsch, Stephan
JOURNAL OF MACHINE LEARNING RESEARCH, 2024, 25
[4] DEEP NEURAL NETWORKS WITH RELU-SINE-EXPONENTIAL ACTIVATIONS BREAK CURSE OF DIMENSIONALITY IN APPROXIMATION ON HOLDER CLASS
Jiao, Yuling
Lai, Yanming
Lu, Xiliang
Wang, Fengru
Yang, Jerry zhijian
Yang, Yuanyuan
SIAM JOURNAL ON MATHEMATICAL ANALYSIS, 2023, 55 (04) : 3635 - 3649
[5] Theory I: Deep networks and the curse of dimensionality
Poggio, T.
Liao, Q.
BULLETIN OF THE POLISH ACADEMY OF SCIENCES-TECHNICAL SCIENCES, 2018, 66 (06) : 761 - 773
[6] Rectified deep neural networks overcome the curse of dimensionality for nonsmooth value functions in zero-sum games of nonlinear stiff systems
Reisinger, Christoph
Zhang, Yufei
ANALYSIS AND APPLICATIONS, 2020, 18 (06) : 951 - 999
[7] A proof that rectified deep neural networks overcome the curse of dimensionality in the numerical approximation of semilinear heat equations
Hutzenthaler, Martin
Jentzen, Arnulf
Kruse, Thomas
Nguyen, Tuan Anh
PARTIAL DIFFERENTIAL EQUATIONS AND APPLICATIONS, 2020, 1 (02):
[8] TRANSFORMING MULTIDIMENSIONAL DATA INTO IMAGES TO OVERCOME THE CURSE OF DIMENSIONALITY
Leygonie, Rebecca
Lobry, Sylvain
Vimont, Guillaume
Wendling, Laurent
2023 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, ICIP, 2023, : 700 - 704
[9] Rectified deep neural networks overcome the curse of dimensionality when approximating solutions of McKean-Vlasov stochastic differential equations ☆
Neufeld, Ariel
Nguyen, Tuan Anh
JOURNAL OF MATHEMATICAL ANALYSIS AND APPLICATIONS, 2025, 541 (01)
[10] Overcoming the curse of dimensionality for approximating Lyapunov functions with deep neural networks under a small-gain condition
Gruene, Lars
IFAC PAPERSONLINE, 2021, 54 (09): : 317 - 322

← 1 2 3 4 5 →