Parallel-in-Time Solutions with Random Projection Neural Networks

Marta M. Betcke, Lisa Maria Kreusser, Davide Murari
8/19/2024
math.NAcs.LG

No official code found. Best implementation: davidemurari/RPNNHybridParareal (3 stars, last commit 2026-02-26).

Abstract

This paper considers one of the fundamental parallel-in-time methods for the solution of ordinary differential equations, Parareal, and extends it by adopting a neural network as a coarse propagator. We provide a theoretical analysis of the convergence properties of the proposed algorithm and show its effectiveness for several examples, including Lorenz and Burgers' equations. In our numerical simulations, we further specialize the underpinning neural architecture to Random Projection Neural Networks (RPNNs), a 2-layer neural network where the first layer weights are drawn at random rather than optimized. This restriction substantially increases the efficiency of fitting RPNN's weights in comparison to a standard feedforward network without negatively impacting the accuracy, as demonstrated in the SIR system example.

View on arXivView PDF

Code Implementations(1)

davidemurari/RPNNHybridPararealCommunity100%

Repository for the paper "Parallel-in-Time Solutions with Random Projection Neural Networks"

31Shell, PythonJun 5, 20247 months ago

Similar Papers

Linear Stability Analysis of Physics-Informed Random Projection Neural Networks for ODEs

84% match2024

RandNet-Parareal: a time-parallel PDE solver using Random Neural Networks

82% match1 repo12026

Random Projection Neural Networks of Best Approximation: Convergence theory and practical applications

81% match2024

Quasi-Random Physics-informed Neural Networks

80% match2025

A Structure-Preserving Framework for Solving Parabolic Partial Differential Equations with Neural Networks

80% match2025

HEATNETs: Explainable Random Feature Neural Networks for High-Dimensional Parabolic PDEs

80% match2025

Cite this paper

@article{betcke2024parallelintime,
  title  = {Parallel-in-Time Solutions with Random Projection Neural Networks},
  author = {Marta M. Betcke and Lisa Maria Kreusser and Davide Murari},
  year   = {2024},
  eprint = {2408.09756},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2408.09756v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2408.09756.svg)](https://codeofpaper.com/paper/2408.09756)

Discussion