Localized PCA-Net Neural Operators for Scalable Solution Reconstruction of Elliptic PDEs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Dhingra, Mrigank, Maulik, Romit, Rasheed, Adil, San, Omer
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918145747320832
author Dhingra, Mrigank
Maulik, Romit
Rasheed, Adil
San, Omer
author_facet Dhingra, Mrigank
Maulik, Romit
Rasheed, Adil
San, Omer
contents Neural operator learning has emerged as a powerful approach for solving partial differential equations (PDEs) in a data-driven manner. However, applying principal component analysis (PCA) to high-dimensional solution fields incurs significant computational overhead. To address this, we propose a patch-based PCA-Net framework that decomposes the solution fields into smaller patches, applies PCA within each patch, and trains a neural operator in the reduced PCA space. We investigate two different patch-based approaches that balance computational efficiency and reconstruction accuracy: (1) local-to-global patch PCA, and (2) local-to-local patch PCA. The trade-off between computational cost and accuracy is analyzed, highlighting the advantages and limitations of each approach. Furthermore, within each approach, we explore two refinements for the most computationally efficient method: (i) introducing overlapping patches with a smoothing filter and (ii) employing a two-step process with a convolutional neural network (CNN) for refinement. Our results demonstrate that patch-based PCA significantly reduces computational complexity while maintaining high accuracy, reducing end-to-end pipeline processing time by a factor of 3.7 to 4 times compared to global PCA, thefore making it a promising technique for efficient operator learning in PDE-based systems.
format Preprint
id arxiv_https___arxiv_org_abs_2509_18110
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Localized PCA-Net Neural Operators for Scalable Solution Reconstruction of Elliptic PDEs
Dhingra, Mrigank
Maulik, Romit
Rasheed, Adil
San, Omer
Machine Learning
Computer Vision and Pattern Recognition
Neural operator learning has emerged as a powerful approach for solving partial differential equations (PDEs) in a data-driven manner. However, applying principal component analysis (PCA) to high-dimensional solution fields incurs significant computational overhead. To address this, we propose a patch-based PCA-Net framework that decomposes the solution fields into smaller patches, applies PCA within each patch, and trains a neural operator in the reduced PCA space. We investigate two different patch-based approaches that balance computational efficiency and reconstruction accuracy: (1) local-to-global patch PCA, and (2) local-to-local patch PCA. The trade-off between computational cost and accuracy is analyzed, highlighting the advantages and limitations of each approach. Furthermore, within each approach, we explore two refinements for the most computationally efficient method: (i) introducing overlapping patches with a smoothing filter and (ii) employing a two-step process with a convolutional neural network (CNN) for refinement. Our results demonstrate that patch-based PCA significantly reduces computational complexity while maintaining high accuracy, reducing end-to-end pipeline processing time by a factor of 3.7 to 4 times compared to global PCA, thefore making it a promising technique for efficient operator learning in PDE-based systems.
title Localized PCA-Net Neural Operators for Scalable Solution Reconstruction of Elliptic PDEs
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2509.18110