{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,5]],"date-time":"2026-05-05T21:13:22Z","timestamp":1778015602073,"version":"3.51.4"},"reference-count":43,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,3,30]],"date-time":"2026-03-30T00:00:00Z","timestamp":1774828800000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000993","name":"Battelle","doi-asserted-by":"publisher","award":["DE-AC05-76RL01830"],"award-info":[{"award-number":["DE-AC05-76RL01830"]}],"id":[{"id":"10.13039\/100000993","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006132","name":"Office of Science","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006132","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100020879","name":"Society for Education in Anesthesia","doi-asserted-by":"publisher","award":["80278"],"award-info":[{"award-number":["80278"]}],"id":[{"id":"10.13039\/100020879","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100011661","name":"Pacific Northwest National Laboratory","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100011661","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000015","name":"U.S. Department of Energy","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000015","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006192","name":"Advanced Scientific Computing Research","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100006192","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Journal of Computational Physics"],"published-print":{"date-parts":[[2026,8]]},"DOI":"10.1016\/j.jcp.2026.114851","type":"journal-article","created":{"date-parts":[[2026,3,23]],"date-time":"2026-03-23T11:39:35Z","timestamp":1774265975000},"page":"114851","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["What do physics-informed DeepONets learn? Understanding and improving training for scientific computing applications"],"prefix":"10.1016","volume":"558","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8379-2570","authenticated-orcid":false,"given":"Emily","family":"Williams","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6411-6198","authenticated-orcid":false,"given":"Amanda","family":"Howard","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0829-3907","authenticated-orcid":false,"given":"Saad","family":"Qadeer","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4683-6578","authenticated-orcid":false,"given":"Brek","family":"Meuris","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9928-5637","authenticated-orcid":false,"given":"Panos","family":"Stinis","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"78","reference":[{"issue":"1","key":"10.1016\/j.jcp.2026.114851_bib0001","doi-asserted-by":"crossref","first-page":"1739","DOI":"10.1038\/s41598-022-26602-3","article-title":"Machine-learning-based spectral methods for partial differential equations","volume":"13","author":"Meuris","year":"2023","journal-title":"Sci. Rep."},{"issue":"6","key":"10.1016\/j.jcp.2026.114851_bib0002","doi-asserted-by":"crossref","first-page":"422","DOI":"10.1038\/s42254-021-00314-5","article-title":"Physics-informed machine learning","volume":"3","author":"Karniadakis","year":"2021","journal-title":"Nat. Rev. Phys."},{"issue":"1","key":"10.1016\/j.jcp.2026.114851_bib0003","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41746-019-0193-y","article-title":"Integrating machine learning and multiscale modeling\u2014perspectives, challenges, and opportunities in the biological, biomedical, and behavioral sciences","volume":"2","author":"Alber","year":"2019","journal-title":"NPJ. Digit. Med."},{"issue":"3","key":"10.1016\/j.jcp.2026.114851_bib0004","doi-asserted-by":"crossref","first-page":"218","DOI":"10.1038\/s42256-021-00302-5","article-title":"Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators","volume":"3","author":"Lu","year":"2021","journal-title":"Nat. Mach. Intell."},{"issue":"40","key":"10.1016\/j.jcp.2026.114851_bib0005","doi-asserted-by":"crossref","DOI":"10.1126\/sciadv.abi8605","article-title":"Learning the solution operator of parametric partial differential equations with physics-informed deepONets","volume":"7","author":"Wang","year":"2021","journal-title":"Sci. Adv."},{"key":"10.1016\/j.jcp.2026.114851_bib0006","unstructured":"S. Qadeer, A. Engel, A. Howard, A. Tsou, M. Vargas, P. Stinis, T. Chiang, Efficient Kernel Surrogates for Neural Network-Based Regression, 2024, 10.48550\/arXiv.2310.18612."},{"issue":"2","key":"10.1016\/j.jcp.2026.114851_bib0007","doi-asserted-by":"crossref","DOI":"10.1007\/s10915-022-01881-0","article-title":"Improved architectures and training algorithms for deep operator networks","volume":"92","author":"Wang","year":"2022","journal-title":"J. Sci. Comput."},{"issue":"Volume 25, 1993","key":"10.1016\/j.jcp.2026.114851_bib0008","doi-asserted-by":"crossref","first-page":"539","DOI":"10.1146\/annurev.fl.25.010193.002543","article-title":"The proper orthogonal decomposition in the analysis of turbulent flows","volume":"25","author":"Berkooz","year":"1993","journal-title":"Annu. Rev. Fluid Mech."},{"issue":"5","key":"10.1016\/j.jcp.2026.114851_bib0009","doi-asserted-by":"crossref","first-page":"1893","DOI":"10.1137\/S0036142901389049","article-title":"A new look at proper orthogonal decomposition","volume":"41","author":"Rathinam","year":"2003","journal-title":"SIAM J. Numer. Anal."},{"key":"10.1016\/j.jcp.2026.114851_bib0010","doi-asserted-by":"crossref","first-page":"395","DOI":"10.1007\/s11831-011-9064-7","article-title":"A short review on model order reduction based on proper generalized decomposition","volume":"18","author":"Chinesta","year":"2011","journal-title":"Arch. Comput. Methods Eng."},{"issue":"2","key":"10.1016\/j.jcp.2026.114851_bib0011","doi-asserted-by":"crossref","first-page":"492","DOI":"10.1137\/S0036142900382612","article-title":"Galerkin proper orthogonal decomposition methods for a general equation in fluid dynamics","volume":"40","author":"Kunisch","year":"2002","journal-title":"SIAM J. Numer. Anal."},{"issue":"7","key":"10.1016\/j.jcp.2026.114851_bib0012","first-page":"808","article-title":"An introduction to the proper orthogonal decomposition","volume":"78","author":"Chatterjee","year":"2000","journal-title":"Curr. Sci."},{"issue":"4","key":"10.1016\/j.jcp.2026.114851_bib0013","doi-asserted-by":"crossref","first-page":"911","DOI":"10.1109\/72.392253","article-title":"Universal approximation to nonlinear operators by neural networks with arbitrary activation functions and its application to dynamical systems","volume":"6","author":"Chen","year":"1995","journal-title":"IEEE Trans. Neural Networks"},{"key":"10.1016\/j.jcp.2026.114851_bib0014","series-title":"Deep Learning","author":"Goodfellow","year":"2016"},{"key":"10.1016\/j.jcp.2026.114851_bib0015","first-page":"121","article-title":"The influence of pattern similarity and transfer learning upon training of a base perceptron b2","volume":"3","author":"Bozinovski","year":"1976","journal-title":"Proc. Symp. Inform."},{"key":"10.1016\/j.jcp.2026.114851_bib0016","article-title":"Transfer learning for text classification","author":"Do","year":"2005","journal-title":"Adv. Neural. Inf. Process. Syst."},{"issue":"5","key":"10.1016\/j.jcp.2026.114851_bib0017","doi-asserted-by":"crossref","first-page":"1285","DOI":"10.1109\/TMI.2016.2528162","article-title":"Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning","volume":"35","author":"Shin","year":"2016","journal-title":"IEEE Trans. Med. Imaging"},{"key":"10.1016\/j.jcp.2026.114851_bib0018","doi-asserted-by":"crossref","unstructured":"A. Conneau, D. Kiela, H. Schwenk, L. Barrault, A. Bordes, Supervised Learning of Universal Sentence Representations from Natural Language Inference Data, 2018, 10.48550\/arXiv.1705.02364.","DOI":"10.18653\/v1\/D17-1070"},{"key":"10.1016\/j.jcp.2026.114851_bib0019","series-title":"2019 IEEE International Conference on Networking, Architecture and Storage (NAS)","first-page":"1","article-title":"Exploring transfer learning to reduce training overhead of HPC data in machine learning","author":"Liu","year":"2019"},{"issue":"2","key":"10.1016\/j.jcp.2026.114851_bib0020","doi-asserted-by":"crossref","first-page":"40","DOI":"10.3390\/technologies11020040","article-title":"A review of deep transfer learning and recent advancements","volume":"11","author":"Iman","year":"2023","journal-title":"Technologies"},{"issue":"1","key":"10.1016\/j.jcp.2026.114851_bib0021","doi-asserted-by":"crossref","first-page":"139","DOI":"10.60087\/jklst.vol1.n1.p147","article-title":"Harnessing the power of transfer learning in deep learning models","volume":"1","author":"Venkatasubbu","year":"2023","journal-title":"J. Knowl. Learn. Sci. Technol. ISSN: 2959\u20136386 (Online)"},{"key":"10.1016\/j.jcp.2026.114851_bib0022","series-title":"Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics","first-page":"249","article-title":"Understanding the difficulty of training deep feedforward neural networks","author":"Glorot","year":"2010"},{"key":"10.1016\/j.jcp.2026.114851_bib0023","series-title":"Advances in Neural Information Processing Systems","first-page":"3320","article-title":"How transferable are features in deep neural networks?","volume":"27","author":"Yosinski","year":"2014"},{"key":"10.1016\/j.jcp.2026.114851_bib0024","doi-asserted-by":"crossref","unstructured":"Z. Zhang, C. Moya, L. Lu, G. Lin, H. Schaeffer, DeepONet as a Multi-Operator Extrapolation Model: Distributed Pretraining with Physics-Informed Fine-Tuning, 2024, https:\/\/arxiv.org\/abs\/2411.07239.","DOI":"10.2139\/ssrn.5223161"},{"key":"10.1016\/j.jcp.2026.114851_bib0025","doi-asserted-by":"crossref","DOI":"10.1016\/j.engappai.2024.108156","article-title":"A multi-fidelity deep operator network (DeepONet) for fusing simulation and monitoring data: application to real-time settlement prediction during tunnel construction","volume":"133","author":"Xu","year":"2024","journal-title":"Eng. Appl. Artif. Intell."},{"issue":"9","key":"10.1016\/j.jcp.2026.114851_bib0026","first-page":"10629","article-title":"Transfer learning enhanced deepONet for long-time prediction of evolution equations","volume":"37","author":"Xu","year":"2023","journal-title":"Proc. of the AAAI Conf. on Artif. Intell."},{"issue":"12","key":"10.1016\/j.jcp.2026.114851_bib0027","doi-asserted-by":"crossref","first-page":"1155","DOI":"10.1038\/s42256-022-00569-2","article-title":"Deep transfer operator learning for partial differential equations under conditional shift","volume":"4","author":"Goswami","year":"2022","journal-title":"Nat. Mach. Intell."},{"key":"10.1016\/j.jcp.2026.114851_bib0028","doi-asserted-by":"crossref","unstructured":"S. Wang, P. Perdikaris, Long-Time Integration of Parametric Evolution Equations with Physics-Informed DeepONets, 2021, 10.48550\/arXiv.2106.05384.","DOI":"10.26226\/morressier.612f6736bc98103724100850"},{"key":"10.1016\/j.jcp.2026.114851_bib0029","doi-asserted-by":"crossref","unstructured":"V. Kumar, S. Goswami, K. Kontolati, M.D. Shields, G.E. Karniadakis, Synergistic Learning with Multi-Task DeepONet for Efficient PDE Problem Solving, 2024, arXiv: 2408.02198.","DOI":"10.1016\/j.neunet.2024.107113"},{"issue":"3","key":"10.1016\/j.jcp.2026.114851_bib0030","doi-asserted-by":"crossref","first-page":"471","DOI":"10.1007\/s00466-023-02271-5","article-title":"On the geometry transferability of the hybrid iterative numerical solver for differential equations","volume":"72","author":"Kahana","year":"2023","journal-title":"Comput. Mech."},{"key":"10.1016\/j.jcp.2026.114851_bib0031","doi-asserted-by":"crossref","unstructured":"L. Hu, L. Chen, Y. Li, X. Liu, G. Chen, Q. Meng, X. Hao, C. Liu, Operator Transfer Learning for Physics Field Prediction on Complex Geometries with Limited Labelled Data, 2024, 10.2139\/ssrn.4981813.","DOI":"10.2139\/ssrn.4981813"},{"key":"10.1016\/j.jcp.2026.114851_bib0032","doi-asserted-by":"crossref","DOI":"10.1016\/j.neunet.2024.106703","article-title":"Kolmogorov n-widths for multitask physics-informed machine learning (PIML) methods: towards robust metrics","volume":"180","author":"Penwarden","year":"2024","journal-title":"Neural Netw."},{"key":"10.1016\/j.jcp.2026.114851_bib0033","doi-asserted-by":"crossref","unstructured":"S. Wu, Fine-Tuning DeepONets to Enhance Physics-informed Neural Networks for Solving Partial Differential Equations, 2024, 10.48550\/arXiv.2410.14134.","DOI":"10.2139\/ssrn.4992639"},{"key":"10.1016\/j.jcp.2026.114851_bib0034","series-title":"Advances in Neural Information Processing Systems","first-page":"8580","article-title":"Neural tangent kernel: convergence and generalization in neural networks","volume":"31","author":"Jacot","year":"2018"},{"key":"10.1016\/j.jcp.2026.114851_bib0035","series-title":"Proceedings of the 36th International Conference on Machine Learning","first-page":"1675","article-title":"Gradient descent finds global minima of deep neural networks","author":"Du","year":"2019"},{"key":"10.1016\/j.jcp.2026.114851_bib0036","series-title":"Proceedings of the 36th International Conference on Machine Learning","first-page":"242","article-title":"A convergence theory for deep learning via over-parameterization","author":"Allen-Zhu","year":"2019"},{"key":"10.1016\/j.jcp.2026.114851_bib0037","series-title":"Advances in Neural Information Processing Systems","first-page":"7710","article-title":"Spectra of the conjugate kernel and neural tangent kernel for linear-width neural networks","volume":"33","author":"Fan","year":"2020"},{"key":"10.1016\/j.jcp.2026.114851_bib0038","series-title":"ICLR 2024 Workshop on AI4DifferentialEquations in Science","first-page":"1","article-title":"The conjugate kernel for efficient training of physics-informed deep operator networks","author":"Howard","year":"2024"},{"key":"10.1016\/j.jcp.2026.114851_bib0039","unstructured":"Z. Li, N. Kovachki, K. Azizzadenesheli, B. Liu, K. Bhattacharya, A. Stuart, A. Anandkumar, Fourier Neural Operator for Parametric Partial Differential Equations, 2021, 10.48550\/arXiv.2010.08895."},{"key":"10.1016\/j.jcp.2026.114851_bib0040","article-title":"Stacked networks improve physics-informed training: applications to neural networks and deep operator networks","author":"Howard","year":"2024","journal-title":"Found. Data Sci."},{"key":"10.1016\/j.jcp.2026.114851_bib0041","unstructured":"E. Kharazmi, Z. Zhang, G.E. Karniadakis, Variational Physics-Informed Neural Networks For Solving Partial Differential Equations, 2019, arXiv: 1912.00873."},{"key":"10.1016\/j.jcp.2026.114851_bib0042","doi-asserted-by":"crossref","DOI":"10.1016\/j.cma.2024.116813","article-title":"Respecting causality for training physics-informed neural networks","volume":"421","author":"Wang","year":"2024","journal-title":"Comput. Methods Appl. Mech. Eng."},{"issue":"123","key":"10.1016\/j.jcp.2026.114851_bib0043","doi-asserted-by":"crossref","first-page":"579","DOI":"10.1090\/S0025-5718-1973-0348991-3","article-title":"Numerical methods for computing angles between linear subspaces","volume":"27","author":"Bj\u00f6rck","year":"1973","journal-title":"Math. Comput."}],"container-title":["Journal of Computational Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0021999126002019?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0021999126002019?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,5,5]],"date-time":"2026-05-05T20:43:18Z","timestamp":1778013798000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0021999126002019"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8]]},"references-count":43,"alternative-id":["S0021999126002019"],"URL":"https:\/\/doi.org\/10.1016\/j.jcp.2026.114851","relation":{},"ISSN":["0021-9991"],"issn-type":[{"value":"0021-9991","type":"print"}],"subject":[],"published":{"date-parts":[[2026,8]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"What do physics-informed DeepONets learn? Understanding and improving training for scientific computing applications","name":"articletitle","label":"Article Title"},{"value":"Journal of Computational Physics","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.jcp.2026.114851","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 The Author(s) and Battelle Memorial Institute. Published by Elsevier Inc.","name":"copyright","label":"Copyright"}],"article-number":"114851"}}