{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,5]],"date-time":"2026-05-05T12:05:20Z","timestamp":1777982720456,"version":"3.51.4"},"reference-count":42,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T00:00:00Z","timestamp":1772496000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T00:00:00Z","timestamp":1772496000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100006260","name":"Technion - Israel Institute of Technology","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100006260","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Quantum Mach. Intell."],"published-print":{"date-parts":[[2026,6]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>\n                    Physics-Informed Neural Networks (PINNs) have emerged as a promising approach for solving partial differential equations (PDEs) by embedding the governing physics into the loss function associated with a deep neural network. In this work, a\n                    <jats:italic>Quantum Physics-Informed Neural Network<\/jats:italic>\n                    (QPINN) framework is proposed to solve two-dimensional (2D) time-dependent Maxwell\u2019s equations. Our approach utilizes a parametrized quantum circuit (PQC) in conjunction with the classical neural network architecture and enforces physical laws, including a global energy conservation principle, during training. A quantum simulation library,\u00a0\n                    <jats:italic>TorQ - Tensor Operations for Research in Quantum systems<\/jats:italic>\n                    , was developed to efficiently compute circuit outputs and derivatives by leveraging GPU acceleration based on PyTorch, enabling end-to-end training of the QPINN. The method was evaluated on two 2D electromagnetic wave propagation problems: one in free space (vacuum) and the other has an added dielectric medium. Multiple quantum circuit ans\u00e4tze, input scales, and an added loss term were compared in a thorough ablation study. Furthermore, recent techniques to enhance PINN convergence, including random Fourier feature embeddings and adaptive time weighting, have been incorporated. Our results demonstrate that the QPINN achieves accuracy comparable to, and even greater than, the classical PINN baseline, while using a significantly smaller number of trainable parameters. This study also demonstrates that adding an energy conservation term to the loss stabilizes training and improves the physical fidelity of the solution in the lossless free-space case. This added term helps mitigate a new kind of barren plateau (BP) related phenomenon -\n                    <jats:italic>\u201cblack hole\u201d<\/jats:italic>\n                    (BH) loss landscape for the quantum experiments in that scenario. By optimizing the quantum-circuit ansatz and embedding energy-conservation constraints, our QPINN achieves up to a\n                    <jats:inline-formula>\n                      <jats:tex-math>$$19\\%$$<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    higher accuracy on 2D Maxwell benchmark problems compared to a classical PINN.\n                  <\/jats:p>","DOI":"10.1007\/s42484-026-00365-w","type":"journal-article","created":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T13:53:45Z","timestamp":1772546025000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Quantum physics-informed neural networks for Maxwell\u2019s equations: circuit design, \u201cblack hole\u201d barren plateaus mitigation, and GPU acceleration"],"prefix":"10.1007","volume":"8","author":[{"given":"Ziv","family":"Chen","sequence":"first","affiliation":[]},{"given":"Gal G.","family":"Shaviner","sequence":"additional","affiliation":[]},{"given":"Hemanth","family":"Chandravamsi","sequence":"additional","affiliation":[]},{"given":"Shimon","family":"Pisnoy","sequence":"additional","affiliation":[]},{"given":"Steven H.","family":"Frankel","sequence":"additional","affiliation":[]},{"given":"Uzi","family":"Pereg","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2026,3,3]]},"reference":[{"key":"365_CR1","doi-asserted-by":"publisher","first-page":"116805","DOI":"10.1016\/j.cma.2024.116805","volume":"421","author":"SJ Anagnostopoulos","year":"2024","unstructured":"Anagnostopoulos SJ, Toscano JD, Stergiopulos N, Karniadakis GE (2024) Residual-based attention in physics-informed neural networks. Comput Methods Appl Mech Eng 421:116805","journal-title":"Comput Methods Appl Mech Eng"},{"key":"365_CR2","doi-asserted-by":"crossref","unstructured":"Barmada S, Dodge S, Tucci M, Formisano A, Di Barba P, Mognaschi ME (2024) A novel hybrid boundary element\u2013physics informed neural network method for numerical solutions in electromagnetics. IEEE Access","DOI":"10.1109\/ACCESS.2024.3500039"},{"key":"365_CR3","unstructured":"Bowles J, Ahmed S, Schuld M (2024) Better than classical? the subtle art of benchmarking quantum machine learning models. arXiv:2403.07059"},{"key":"365_CR4","unstructured":"Brendel P, Rosskopf A, Medvedev V (2022) Physics-informed neural networks for quasi-static maxwell\u2019s equations, Fraunhofer IISB Erlangen"},{"issue":"1","key":"365_CR5","doi-asserted-by":"publisher","first-page":"1791","DOI":"10.1038\/s41467-021-21728-w","volume":"12","author":"M Cerezo","year":"2021","unstructured":"Cerezo M, Sone A, Volkoff T, Cincio L, Coles PJ (2021) Cost function dependent barren plateaus in shallow parametrized quantum circuits. Nat Commun 12(1):1791","journal-title":"Nat Commun"},{"key":"365_CR6","unstructured":"Chang SY, Cerezo M (2025) A primer on quantum machine learning. arXiv:2511.15969"},{"issue":"8","key":"365_CR7","doi-asserted-by":"publisher","first-page":"11618","DOI":"10.1364\/OE.384875","volume":"28","author":"Y Chen","year":"2020","unstructured":"Chen Y, Lu L, Karniadakis GE, Dal Negro L (2020) Physics-informed neural networks for inverse problems in nano-optics and metamaterials. Opt Express 28(8):11618\u201311633","journal-title":"Opt Express"},{"key":"365_CR8","doi-asserted-by":"publisher","first-page":"110242","DOI":"10.1016\/j.jcp.2021.110242","volume":"435","author":"S Dong","year":"2021","unstructured":"Dong S, Ni N (2021) A method for representing periodic functions and enforcing exactly periodic boundary conditions with deep neural networks. J Comput Phys 435:110242","journal-title":"J Comput Phys"},{"key":"365_CR9","doi-asserted-by":"publisher","first-page":"113741","DOI":"10.1016\/j.cma.2021.113741","volume":"379","author":"E Haghighat","year":"2021","unstructured":"Haghighat E, Raissi M, Moure A, Gomez H, Juanes R (2021) A physics-informed deep learning framework for inversion and surrogate modeling in solid mechanics. Comput Methods Appl Mech Eng 379:113741","journal-title":"Comput Methods Appl Mech Eng"},{"key":"365_CR10","doi-asserted-by":"crossref","unstructured":"Zhongkai H, Yao J, Su C, Su H, Wang Z, Lu F,\u00a0Zeyu X, Yichi Z, Songming L, Lu L, Jun Z\u00a0(2023) Pinnacle: A comprehensive benchmark of physics-informed neural networks for solving pdes.\u00a0Adv Neural Inf Process Syst\u00a037:76721\u201376774","DOI":"10.52202\/079017-2442"},{"key":"365_CR11","doi-asserted-by":"publisher","first-page":"109951","DOI":"10.1016\/j.jcp.2020.109951","volume":"426","author":"X Jin","year":"2021","unstructured":"Jin X, Cai S, Li H, Karniadakis GE (2021) Nsfnets (navier-stokes flow nets): Physics-informed neural networks for the incompressible navier-stokes equations. J Comput Phys 426:109951","journal-title":"J Comput Phys"},{"key":"365_CR12","unstructured":"Kingma DP, Ba J (2014) Adam: A method for stochastic optimization. arXiv:1412.6980"},{"key":"365_CR13","first-page":"26548","volume":"34","author":"A Krishnapriyan","year":"2021","unstructured":"Krishnapriyan A, Gholami A, Zhe S, Kirby R, Mahoney MW (2021) Characterizing possible failure modes in physics-informed neural networks. Adv Neural Inf Process Syst 34:26548\u201326560","journal-title":"Adv Neural Inf Process Syst"},{"key":"365_CR14","doi-asserted-by":"publisher","first-page":"174","DOI":"10.1038\/s42254-025-00813-9","volume":"7","author":"M Larocca","year":"2025","unstructured":"Larocca M, Thanasilp S, Wang S, Sharma K, Biamonte J, Coles PJ, Cincio L, McClean JR, Holmes Z, Cerezo M (2025) Barren plateaus in variational quantum computing. Nat Rev Phys 7:174\u2013189","journal-title":"Nat Rev Phys"},{"key":"365_CR15","doi-asserted-by":"crossref","unstructured":"Larocca M, Thanasilp S, Wang S, Sharma K, Biamonte J, Coles PJ, Cincio L, McClean JR, Holmes Z, Cerezo M (2025) Barren plateaus in variational quantum computing. Nat Rev Phys 1\u201316","DOI":"10.1038\/s42254-025-00813-9"},{"key":"365_CR16","doi-asserted-by":"crossref","unstructured":"Lim J, Psaltis D (2022) Maxwellnet: Physics-driven deep neural network training based on maxwell\u2019s equations. A Photonics 7(1)","DOI":"10.1063\/5.0071616"},{"issue":"1","key":"365_CR17","doi-asserted-by":"publisher","first-page":"015058","DOI":"10.1088\/2632-2153\/ad35a3","volume":"5","author":"J Liu","year":"2024","unstructured":"Liu J, Lin Z, Jiang L (2024) Laziness, barren plateau, and noises in machine learning. Mach Learn Sci Technol 5(1):015058","journal-title":"Mach Learn Sci Technol"},{"issue":"1","key":"365_CR18","doi-asserted-by":"publisher","first-page":"4812","DOI":"10.1038\/s41467-018-07090-4","volume":"9","author":"JR McClean","year":"2018","unstructured":"McClean JR, Boixo S, Smelyanskiy VN, Babbush R, Neven H (2018) Barren plateaus in quantum neural network training landscapes. Nat Commun 9(1):4812","journal-title":"Nat Commun"},{"key":"365_CR19","doi-asserted-by":"publisher","unstructured":"Meyer DA, Wallach NR (2002) Global entanglement in multiparticle systems. J Math Phys 43(9):4273\u20134278. https:\/\/doi.org\/10.1063\/1.1497700","DOI":"10.1063\/1.1497700"},{"issue":"3","key":"365_CR20","first-page":"032309","volume":"98","author":"K Mitarai","year":"2018","unstructured":"Mitarai K, Negoro M, Kitagawa M, Fujii K (2018) Quant Circ Learn Phys Rev A 98(3):032309","journal-title":"Quant Circ Learn Phys Rev A"},{"key":"365_CR21","doi-asserted-by":"crossref","unstructured":"Nemirovsky J, Chuchem M, Shapira Y (2025) Efficient compilation of quantum circuits using multi-qubit gates. Quantum Sci Technol","DOI":"10.1088\/2058-9565\/ae36cc"},{"key":"365_CR22","doi-asserted-by":"publisher","first-page":"404","DOI":"10.1109\/OJAP.2020.3013830","volume":"1","author":"O Noakoasteen","year":"2020","unstructured":"Noakoasteen O, Wang S, Peng Z, Christodoulou C (2020) Physics-informed deep neural networks for transient electromagnetic analysis. IEEE Open J Anten Propa 1:404\u2013412","journal-title":"IEEE Open J Anten Propa"},{"key":"365_CR23","doi-asserted-by":"crossref","unstructured":"Panichi G, Corli S, Prati E (2026) Quantum physics-informed neural networks for multivariable partial differential equations.\u00a0Phys Rev Appl 25(1):014001","DOI":"10.1103\/6nh4-yh2y"},{"key":"365_CR24","doi-asserted-by":"publisher","first-page":"226","DOI":"10.22331\/q-2020-02-06-226","volume":"4","author":"A P\u00e9rez-Salinas","year":"2020","unstructured":"P\u00e9rez-Salinas A, Cervera-Lierta A, Gil-Fuster E, Latorre JI (2020) Data re-uploading for a universal quantum classifier. Quantum 4:226","journal-title":"Quantum"},{"key":"365_CR25","doi-asserted-by":"crossref","unstructured":"Piao S, Gu H, Wang A, Qin P (2024) A domain-adaptive physics-informed neural network for inverse problems of maxwell\u2019s equations in heterogeneous media. IEEE Antennas and Wireless Propagation Letters","DOI":"10.1109\/LAWP.2024.3413851"},{"key":"365_CR26","unstructured":"Rahimi A, Recht B (2007) Random features for large-scale kernel machines. Adv Neural Inf Process Syst 20"},{"key":"365_CR27","doi-asserted-by":"publisher","first-page":"686","DOI":"10.1016\/j.jcp.2018.10.045","volume":"378","author":"M Raissi","year":"2019","unstructured":"Raissi M, Perdikaris P, Karniadakis GE (2019) Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J Comput Phys 378:686\u2013707","journal-title":"J Comput Phys"},{"key":"365_CR28","doi-asserted-by":"crossref","unstructured":"Rasht-Behesht M, Huber C, Shukla K, Karniadakis GE (2022) Physics-informed neural networks (pinns) for wave propagation and full waveform inversions. J Geophys Res: Solid Earth 127(5):e2021JB023120","DOI":"10.1029\/2021JB023120"},{"issue":"2","key":"365_CR29","doi-asserted-by":"publisher","first-page":"020365","DOI":"10.1103\/PRXQuantum.3.020365","volume":"3","author":"SH Sack","year":"2022","unstructured":"Sack SH, Medina RA, Michailidis AA, Kueng R, Serbyn M (2022) Avoiding barren plateaus using classical shadows. PRX Quantum 3(2):020365","journal-title":"PRX Quantum"},{"issue":"3","key":"365_CR30","first-page":"032308","volume":"101","author":"M Schuld","year":"2020","unstructured":"Schuld M, Bocharov A (2020) Krysta M Svore, and Nathan Wiebe. Circuit-cent Quant Classif Phys Rev A 101(3):032308","journal-title":"Circuit-cent Quant Classif Phys Rev A"},{"issue":"3","key":"365_CR31","doi-asserted-by":"publisher","first-page":"032430","DOI":"10.1103\/PhysRevA.103.032430","volume":"103","author":"M Schuld","year":"2021","unstructured":"Schuld M, Sweke R, Meyer JJ (2021) Effect of data encoding on the expressive power of variational quantum-machine-learning models. Phys Rev A 103(3):032430","journal-title":"Phys Rev A"},{"issue":"2","key":"365_CR32","doi-asserted-by":"publisher","first-page":"025045","DOI":"10.1088\/2632-2153\/ad43b2","volume":"5","author":"A Sedykh","year":"2024","unstructured":"Sedykh A, Podapaka M, Sagingalieva A, Pinto K, Pflitsch M, Melnikov A (2024) Hybrid quantum physics-informed neural networks for simulating computational fluid dynamics in complex shapes. Mach Learn Sci Technol 5(2):025045","journal-title":"Mach Learn Sci Technol"},{"key":"365_CR33","doi-asserted-by":"publisher","unstructured":"Shaviner GG, Chandravamsi H, Pisnoy S et al (2025) PINNs for solving unsteady Maxwell\u2019s equations: convergence issues and comparative assessment with compact schemes. Neural Comput and Applic 37(29):24103\u201324122. https:\/\/doi.org\/10.1007\/s00521-025-11554-2","DOI":"10.1007\/s00521-025-11554-2"},{"issue":"12","key":"365_CR34","doi-asserted-by":"publisher","first-page":"1900070","DOI":"10.1002\/qute.201900070","volume":"2","author":"S Sim","year":"2019","unstructured":"Sim S, Johnson PD, Aspuru-Guzik A (2019) Expressibility and entangling capability of parameterized quantum circuits for hybrid quantum-classical algorithms. Adv Quant Technol 2(12):1900070","journal-title":"Adv Quant Technol"},{"key":"365_CR35","first-page":"7537","volume":"33","author":"M Tancik","year":"2020","unstructured":"Tancik M, Srinivasan P, Mildenhall B, Fridovich-Keil S, Raghavan N, Singhal U, Ramamoorthi R, Barron J, Ng R (2020) Fourier features let networks learn high frequency functions in low dimensional domains. Adv Neural Inf Process Syst 33:7537\u20137547","journal-title":"Adv Neural Inf Process Syst"},{"issue":"1","key":"365_CR36","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s44379-025-00015-1","volume":"1","author":"JD Toscano","year":"2025","unstructured":"Toscano JD, Oommen V, Varghese AJ, Zou Z, Ahmadi Daryakenari N, Wu C, Karniadakis GE (2025) From pinns to pikans: Recent advances in physics-informed machine learning. Mach Learn Comput Sci Eng 1(1):1\u201343","journal-title":"Mach Learn Comput Sci Eng"},{"issue":"8","key":"365_CR37","doi-asserted-by":"publisher","first-page":"649","DOI":"10.3390\/e26080649","volume":"26","author":"C Trahan","year":"2024","unstructured":"Trahan C, Loveland M, Dent S (2024) Quantum physics-informed neural networks. Entropy 26(8):649","journal-title":"Entropy"},{"key":"365_CR38","doi-asserted-by":"publisher","first-page":"113938","DOI":"10.1016\/j.cma.2021.113938","volume":"384","author":"S Wang","year":"2021","unstructured":"Wang S, Wang H, Perdikaris P (2021) On the eigenvector bias of fourier feature networks: From regression to solving multi-scale pdes with physics-informed neural networks. Comput Methods Appl Mech Eng 384:113938","journal-title":"Comput Methods Appl Mech Eng"},{"key":"365_CR39","doi-asserted-by":"crossref","unstructured":"Wang S, Sankaran S, Perdikaris,P (2024) Respecting causality for training physics-informed neural networks.\u00a0Comput Meth Appl Mech Eng 421:116813","DOI":"10.1016\/j.cma.2024.116813"},{"key":"365_CR40","doi-asserted-by":"crossref","unstructured":"Wang S, Sankaran S, Wang H, Perdikaris P (2023) An expert\u2019s guide to training physics-informed neural networks. arXiv:2308.08468","DOI":"10.1016\/j.cma.2024.116813"},{"key":"365_CR41","doi-asserted-by":"publisher","first-page":"111260","DOI":"10.1016\/j.jcp.2022.111260","volume":"462","author":"L Yuan","year":"2022","unstructured":"Yuan L, Ni Y-Q, Deng X-Y, Hao S (2022) A-pinn: Auxiliary physics informed neural networks for forward and inverse problems of nonlinear integro-differential equations. J Comput Phys 462:111260","journal-title":"J Comput Phys"},{"key":"365_CR42","doi-asserted-by":"crossref","unstructured":"Zheng Y-R, Huang Z-Y, Gong X-Z, Zheng X-Z (2024) Implementation of maxwell\u2019s equations solving algorithm based on pinn. In 2024 IEEE International Conference on Computational Electromagnetics (ICCEM), pp 1\u20133. IEEE","DOI":"10.1109\/ICCEM60619.2024.10559130"}],"container-title":["Quantum Machine Intelligence"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s42484-026-00365-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s42484-026-00365-w","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s42484-026-00365-w.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,3]],"date-time":"2026-03-03T13:54:12Z","timestamp":1772546052000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s42484-026-00365-w"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,3]]},"references-count":42,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,6]]}},"alternative-id":["365"],"URL":"https:\/\/doi.org\/10.1007\/s42484-026-00365-w","relation":{"has-preprint":[{"id-type":"doi","id":"10.21203\/rs.3.rs-7010088\/v1","asserted-by":"object"}]},"ISSN":["2524-4906","2524-4914"],"issn-type":[{"value":"2524-4906","type":"print"},{"value":"2524-4914","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,3]]},"assertion":[{"value":"30 June 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"26 January 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"3 March 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare that they have no conflicts of interest.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Conflicts of Interest"}},{"value":"The authors declare no competing interests.","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing Interests"}}],"article-number":"21"}}