Song, Y. orcid.org/0000-0003-4610-027X, Wang, H., Yang, H. orcid.org/0000-0003-4931-0168 et al. (2 more authors) (2024) Loss-attentional physics-informed neural networks. Journal of Computational Physics, 501. 112781. ISSN 0021-9991
Abstract
Physics-informed neural networks (PINNs) have emerged as a significant endeavour in recent years to utilize artificial intelligence technology for solving various partial differential equations (PDEs). Nevertheless, the vanilla PINN model structure encounters challenges in accurately approximating solutions at hard-to-fit regions with, for instance, “stiffness” points characterized by fast-paced alterations in timescale. To this end, we introduce a novel model architecture based on PINN, named loss-attentional physics-informed neural networks (LA-PINN), which equips each loss component with an independent loss-attentional network (LAN). Feeding the squared errors (SE) on every training point into LAN as the input, the attentional function is then built by each LAN and provides different weights to diverse point SEs. A point error-based weighting approach that utilizes the adversarial training between multiple networks in the LA-PINN model is proposed to dynamically update weights of SE during every training epoch. Additionally, the weighting mechanism of LA-PINN is analysed and also be validated by performing several numerical experiments. The experimental results indicate that the proposed method displays superior predictive performance compared to the vanilla PINN and holds a swift convergence characteristic. Moreover, it can advance the convergence of those hard-to-fit points by progressively increasing the growth rates of both the weight and the update gradient for point error.
Metadata
Item Type: | Article |
---|---|
Authors/Creators: |
|
Copyright, Publisher and Additional Information: | © 2024 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). |
Keywords: | Physics-informed neural network; Point error-based weighting method; Loss attention; Novel architecture; Neural network |
Dates: |
|
Institution: | The University of Leeds |
Academic Units: | The University of Leeds > Faculty of Engineering & Physical Sciences (Leeds) > School of Civil Engineering (Leeds) |
Depositing User: | Symplectic Publications |
Date Deposited: | 30 Jan 2024 15:43 |
Last Modified: | 30 Jan 2024 15:43 |
Status: | Published |
Publisher: | Elsevier |
Identification Number: | 10.1016/j.jcp.2024.112781 |
Open Archives Initiative ID (OAI ID): | oai:eprints.whiterose.ac.uk:208404 |