Deep vs. Shallow: Benchmarking Physics-Informed Neural Architectures on the Biharmonic Equation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Srinivasan, Akshay Govind, Dwivedi, Vikas, Srinivasan, Balaji
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915533978337280
author Srinivasan, Akshay Govind
Dwivedi, Vikas
Srinivasan, Balaji
author_facet Srinivasan, Akshay Govind
Dwivedi, Vikas
Srinivasan, Balaji
contents Partial differential equation (PDE) solvers are fundamental to engineering simulation. Classical mesh-based approaches (finite difference/volume/element) are fast and accurate on high-quality meshes but struggle with higher-order operators and complex, hard-to-mesh geometries. Recently developed physics-informed neural networks (PINNs) and their variants are mesh-free and flexible, yet compute-intensive and often less accurate. This paper systematically benchmarks RBF-PIELM, a rapid PINN variant-an extreme learning machine with radial-basis activations-for higher-order PDEs. RBF-PIELM replaces PINNs' time-consuming gradient descent with a single-shot least-squares solve. We test RBF-PIELM on the fourth-order biharmonic equation using two benchmarks: lid-driven cavity flow (streamfunction formulation) and a manufactured oscillatory solution. Our results show up to $(350\times)$ faster training than PINNs and over $(10\times)$ fewer parameters for comparable solution accuracy. Despite surpassing PINNs, RBF-PIELM still lags mature mesh-based solvers and its accuracy degrades on highly oscillatory solutions, highlighting remaining challenges for practical deployment.
format Preprint
id arxiv_https___arxiv_org_abs_2510_04490
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Deep vs. Shallow: Benchmarking Physics-Informed Neural Architectures on the Biharmonic Equation
Srinivasan, Akshay Govind
Dwivedi, Vikas
Srinivasan, Balaji
Computational Engineering, Finance, and Science
Emerging Technologies
Machine Learning
G.1.7; G.1.8; G.1.10; J.2
Partial differential equation (PDE) solvers are fundamental to engineering simulation. Classical mesh-based approaches (finite difference/volume/element) are fast and accurate on high-quality meshes but struggle with higher-order operators and complex, hard-to-mesh geometries. Recently developed physics-informed neural networks (PINNs) and their variants are mesh-free and flexible, yet compute-intensive and often less accurate. This paper systematically benchmarks RBF-PIELM, a rapid PINN variant-an extreme learning machine with radial-basis activations-for higher-order PDEs. RBF-PIELM replaces PINNs' time-consuming gradient descent with a single-shot least-squares solve. We test RBF-PIELM on the fourth-order biharmonic equation using two benchmarks: lid-driven cavity flow (streamfunction formulation) and a manufactured oscillatory solution. Our results show up to $(350\times)$ faster training than PINNs and over $(10\times)$ fewer parameters for comparable solution accuracy. Despite surpassing PINNs, RBF-PIELM still lags mature mesh-based solvers and its accuracy degrades on highly oscillatory solutions, highlighting remaining challenges for practical deployment.
title Deep vs. Shallow: Benchmarking Physics-Informed Neural Architectures on the Biharmonic Equation
topic Computational Engineering, Finance, and Science
Emerging Technologies
Machine Learning
G.1.7; G.1.8; G.1.10; J.2
url https://arxiv.org/abs/2510.04490