ELM-DeepONets: Backpropagation-Free Training of Deep Operator Networks via Extreme Learning Machines

Deep Operator Networks (DeepONets) are among the most prominent frameworks for operator learning, grounded in the universal approximation theorem for operators. However, training DeepONets typically requires significant computational resources. To address this limitation, we propose ELM-DeepONets, a...

Full description

Saved in:
Bibliographic Details
Main Author: Hwijae Son
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11005528/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850149557998977024
author Hwijae Son
author_facet Hwijae Son
author_sort Hwijae Son
collection DOAJ
description Deep Operator Networks (DeepONets) are among the most prominent frameworks for operator learning, grounded in the universal approximation theorem for operators. However, training DeepONets typically requires significant computational resources. To address this limitation, we propose ELM-DeepONets, an Extreme Learning Machine (ELM) framework for DeepONets that leverages the backpropagation-free nature of ELM. By reformulating DeepONet training as a least-squares problem for newly introduced parameters, the ELM-DeepONet approach significantly reduces training complexity. Validation on benchmark problems, including nonlinear ODEs and PDEs, demonstrates that the proposed method not only achieves superior accuracy but also drastically reduces computational costs. This work offers a scalable and efficient alternative for operator learning in scientific computing.
format Article
id doaj-art-c3fe6fc63e2841c3be987744d03727d5
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-c3fe6fc63e2841c3be987744d03727d52025-08-20T02:26:51ZengIEEEIEEE Access2169-35362025-01-0113869278693410.1109/ACCESS.2025.357050211005528ELM-DeepONets: Backpropagation-Free Training of Deep Operator Networks via Extreme Learning MachinesHwijae Son0https://orcid.org/0000-0002-6630-2832Department of Mathematics, Konkuk University, Seoul, South KoreaDeep Operator Networks (DeepONets) are among the most prominent frameworks for operator learning, grounded in the universal approximation theorem for operators. However, training DeepONets typically requires significant computational resources. To address this limitation, we propose ELM-DeepONets, an Extreme Learning Machine (ELM) framework for DeepONets that leverages the backpropagation-free nature of ELM. By reformulating DeepONet training as a least-squares problem for newly introduced parameters, the ELM-DeepONet approach significantly reduces training complexity. Validation on benchmark problems, including nonlinear ODEs and PDEs, demonstrates that the proposed method not only achieves superior accuracy but also drastically reduces computational costs. This work offers a scalable and efficient alternative for operator learning in scientific computing.https://ieeexplore.ieee.org/document/11005528/DeepONetsextreme learning machineforward-inverse problems
spellingShingle Hwijae Son
ELM-DeepONets: Backpropagation-Free Training of Deep Operator Networks via Extreme Learning Machines
IEEE Access
DeepONets
extreme learning machine
forward-inverse problems
title ELM-DeepONets: Backpropagation-Free Training of Deep Operator Networks via Extreme Learning Machines
title_full ELM-DeepONets: Backpropagation-Free Training of Deep Operator Networks via Extreme Learning Machines
title_fullStr ELM-DeepONets: Backpropagation-Free Training of Deep Operator Networks via Extreme Learning Machines
title_full_unstemmed ELM-DeepONets: Backpropagation-Free Training of Deep Operator Networks via Extreme Learning Machines
title_short ELM-DeepONets: Backpropagation-Free Training of Deep Operator Networks via Extreme Learning Machines
title_sort elm deeponets backpropagation free training of deep operator networks via extreme learning machines
topic DeepONets
extreme learning machine
forward-inverse problems
url https://ieeexplore.ieee.org/document/11005528/
work_keys_str_mv AT hwijaeson elmdeeponetsbackpropagationfreetrainingofdeepoperatornetworksviaextremelearningmachines