Rectified deep neural networks overcome the curse of dimensionality when approximating solutions of McKean–Vlasov stochastic differential equations

In this paper we prove that rectified deep neural networks do not suffer from the curse of dimensionality when approximating McKean–Vlasov SDEs in the sense that the number of parameters in the deep neural networks only grows polynomially in the space dimension d of the SDE and the reciprocal of the...

Full description

Saved in:
Bibliographic Details
Main Authors: Neufeld, Ariel, Nguyen, Tuan Anh
Other Authors: School of Physical and Mathematical Sciences
Format: Article
Language:English
Published: 2024
Subjects:
Online Access:https://hdl.handle.net/10356/180556
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-180556
record_format dspace
spelling sg-ntu-dr.10356-1805562024-10-14T15:35:24Z Rectified deep neural networks overcome the curse of dimensionality when approximating solutions of McKean–Vlasov stochastic differential equations Neufeld, Ariel Nguyen, Tuan Anh School of Physical and Mathematical Sciences Mathematical Sciences McKean–Vlasov SDEs High-dimensional SDEs In this paper we prove that rectified deep neural networks do not suffer from the curse of dimensionality when approximating McKean–Vlasov SDEs in the sense that the number of parameters in the deep neural networks only grows polynomially in the space dimension d of the SDE and the reciprocal of the accuracy ϵ. Nanyang Technological University Submitted/Accepted version Financial support by the Nanyang Assistant Professorship Grant (NAP Grant) Machine Learning based Algorithms in Finance and Insurance is gratefully acknowledged. 2024-10-11T07:24:14Z 2024-10-11T07:24:14Z 2025 Journal Article Neufeld, A. & Nguyen, T. A. (2025). Rectified deep neural networks overcome the curse of dimensionality when approximating solutions of McKean–Vlasov stochastic differential equations. Journal of Mathematical Analysis and Applications, 541(1), 128661-. https://dx.doi.org/10.1016/j.jmaa.2024.128661 0022-247X https://hdl.handle.net/10356/180556 10.1016/j.jmaa.2024.128661 2-s2.0-85199091793 1 541 128661 en NTU-NAP Journal of Mathematical Analysis and Applications © 2024 Elsevier Inc. All rights reserved. This article may be downloaded for personal use only. Any other use requires prior permission of the copyright holder. The Version of Record is available online at http://doi.org/10.1016/j.jmaa.2024.128661. application/pdf
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Mathematical Sciences
McKean–Vlasov SDEs
High-dimensional SDEs
spellingShingle Mathematical Sciences
McKean–Vlasov SDEs
High-dimensional SDEs
Neufeld, Ariel
Nguyen, Tuan Anh
Rectified deep neural networks overcome the curse of dimensionality when approximating solutions of McKean–Vlasov stochastic differential equations
description In this paper we prove that rectified deep neural networks do not suffer from the curse of dimensionality when approximating McKean–Vlasov SDEs in the sense that the number of parameters in the deep neural networks only grows polynomially in the space dimension d of the SDE and the reciprocal of the accuracy ϵ.
author2 School of Physical and Mathematical Sciences
author_facet School of Physical and Mathematical Sciences
Neufeld, Ariel
Nguyen, Tuan Anh
format Article
author Neufeld, Ariel
Nguyen, Tuan Anh
author_sort Neufeld, Ariel
title Rectified deep neural networks overcome the curse of dimensionality when approximating solutions of McKean–Vlasov stochastic differential equations
title_short Rectified deep neural networks overcome the curse of dimensionality when approximating solutions of McKean–Vlasov stochastic differential equations
title_full Rectified deep neural networks overcome the curse of dimensionality when approximating solutions of McKean–Vlasov stochastic differential equations
title_fullStr Rectified deep neural networks overcome the curse of dimensionality when approximating solutions of McKean–Vlasov stochastic differential equations
title_full_unstemmed Rectified deep neural networks overcome the curse of dimensionality when approximating solutions of McKean–Vlasov stochastic differential equations
title_sort rectified deep neural networks overcome the curse of dimensionality when approximating solutions of mckean–vlasov stochastic differential equations
publishDate 2024
url https://hdl.handle.net/10356/180556
_version_ 1814777740718505984