A comprehensive and FAIR comparison between MLP and KAN representations for differential equations and operator networks

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Shukla, Khemraj, Toscano, Juan Diego, Wang, Zhicheng, Zou, Zongren, Karniadakis, George Em
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910472825995264
author Shukla, Khemraj
Toscano, Juan Diego
Wang, Zhicheng
Zou, Zongren
Karniadakis, George Em
author_facet Shukla, Khemraj
Toscano, Juan Diego
Wang, Zhicheng
Zou, Zongren
Karniadakis, George Em
contents Kolmogorov-Arnold Networks (KANs) were recently introduced as an alternative representation model to MLP. Herein, we employ KANs to construct physics-informed machine learning models (PIKANs) and deep operator models (DeepOKANs) for solving differential equations for forward and inverse problems. In particular, we compare them with physics-informed neural networks (PINNs) and deep operator networks (DeepONets), which are based on the standard MLP representation. We find that although the original KANs based on the B-splines parameterization lack accuracy and efficiency, modified versions based on low-order orthogonal polynomials have comparable performance to PINNs and DeepONet although they still lack robustness as they may diverge for different random seeds or higher order orthogonal polynomials. We visualize their corresponding loss landscapes and analyze their learning dynamics using information bottleneck theory. Our study follows the FAIR principles so that other researchers can use our benchmarks to further advance this emerging topic.
format Preprint
id arxiv_https___arxiv_org_abs_2406_02917
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A comprehensive and FAIR comparison between MLP and KAN representations for differential equations and operator networks
Shukla, Khemraj
Toscano, Juan Diego
Wang, Zhicheng
Zou, Zongren
Karniadakis, George Em
Machine Learning
Computational Physics
Kolmogorov-Arnold Networks (KANs) were recently introduced as an alternative representation model to MLP. Herein, we employ KANs to construct physics-informed machine learning models (PIKANs) and deep operator models (DeepOKANs) for solving differential equations for forward and inverse problems. In particular, we compare them with physics-informed neural networks (PINNs) and deep operator networks (DeepONets), which are based on the standard MLP representation. We find that although the original KANs based on the B-splines parameterization lack accuracy and efficiency, modified versions based on low-order orthogonal polynomials have comparable performance to PINNs and DeepONet although they still lack robustness as they may diverge for different random seeds or higher order orthogonal polynomials. We visualize their corresponding loss landscapes and analyze their learning dynamics using information bottleneck theory. Our study follows the FAIR principles so that other researchers can use our benchmarks to further advance this emerging topic.
title A comprehensive and FAIR comparison between MLP and KAN representations for differential equations and operator networks
topic Machine Learning
Computational Physics
url https://arxiv.org/abs/2406.02917