Saved in:
Bibliographic Details
Main Authors: De Ryck, T., Mishra, S., Shang, Y., Wang, F.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2501.12145
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915113395552256
author De Ryck, T.
Mishra, S.
Shang, Y.
Wang, F.
author_facet De Ryck, T.
Mishra, S.
Shang, Y.
Wang, F.
contents We present approximation results and numerical experiments for the use of randomized neural networks within physics-informed extreme learning machines to efficiently solve high-dimensional PDEs, demonstrating both high accuracy and low computational cost. Specifically, we prove that RaNNs can approximate certain classes of functions, including Sobolev functions, in the $H^2$-norm at dimension-independent convergence rates, thereby alleviating the curse of dimensionality. Numerical experiments are provided for the high-dimensional heat equation, the Black-Scholes model, and the Heston model, demonstrating the accuracy and efficiency of randomized neural networks.
format Preprint
id arxiv_https___arxiv_org_abs_2501_12145
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Approximation Theory and Applications of Randomized Neural Networks for Solving High-Dimensional PDEs
De Ryck, T.
Mishra, S.
Shang, Y.
Wang, F.
Numerical Analysis
We present approximation results and numerical experiments for the use of randomized neural networks within physics-informed extreme learning machines to efficiently solve high-dimensional PDEs, demonstrating both high accuracy and low computational cost. Specifically, we prove that RaNNs can approximate certain classes of functions, including Sobolev functions, in the $H^2$-norm at dimension-independent convergence rates, thereby alleviating the curse of dimensionality. Numerical experiments are provided for the high-dimensional heat equation, the Black-Scholes model, and the Heston model, demonstrating the accuracy and efficiency of randomized neural networks.
title Approximation Theory and Applications of Randomized Neural Networks for Solving High-Dimensional PDEs
topic Numerical Analysis
url https://arxiv.org/abs/2501.12145