{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,2,22]],"date-time":"2025-02-22T00:00:10Z","timestamp":1740182410502,"version":"3.37.3"},"reference-count":95,"publisher":"IOP Publishing","issue":"2","license":[{"start":{"date-parts":[[2024,5,16]],"date-time":"2024-05-16T00:00:00Z","timestamp":1715817600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2024,5,16]],"date-time":"2024-05-16T00:00:00Z","timestamp":1715817600000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"name":"Department Strategic Project of the University of Udine","award":["PSD-AI (2020-2025)"],"award-info":[{"award-number":["PSD-AI (2020-2025)"]}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2024,6,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Federated learning (FL) is an evolving machine learning technique that allows collaborative model training without sharing the original data among participants. In real-world scenarios, data residing at multiple clients are often heterogeneous in terms of different resolutions, magnifications, scanners, or imaging protocols, and thus challenging for global FL model convergence in collaborative training. Most of the existing FL methods consider data heterogeneity within one domain by assuming same data variation in each client site. In this paper, we consider data heterogeneity in FL with different domains of heterogeneous data by raising the problems of domain-shift, class-imbalance, and missing data. We propose a method, multi-domain FL as a solution to heterogeneous training data from multiple domains by training robust vision transformer model. We use two loss functions, one for correctly predicting class labels and other for encouraging similarity and dissimilarity over latent features, to optimize the global FL model. We perform various experiments using different convolution-based networks and non-convolutional Transformer architectures on multi-domain datasets. We evaluate the proposed approach on benchmark datasets and compare with the existing FL methods. Our results show the superiority of the proposed approach which performs better in term of robust FL global model than the exiting methods.<\/jats:p>","DOI":"10.1088\/2632-2153\/ad4768","type":"journal-article","created":{"date-parts":[[2024,5,3]],"date-time":"2024-05-03T22:45:22Z","timestamp":1714776322000},"page":"025041","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Exploiting data diversity in multi-domain federated learning"],"prefix":"10.1088","volume":"5","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1227-524X","authenticated-orcid":true,"given":"Hussain","family":"Ahmad Madni","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Rao","family":"Muhammad Umer","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gian","family":"Luca Foresti","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"266","published-online":{"date-parts":[[2024,5,16]]},"reference":[{"key":"mlstad4768bib1","first-page":"pp 1273","article-title":"Communication-efficient learning of deep networks from decentralized data","author":"McMahan","year":"2017"},{"key":"mlstad4768bib2","first-page":"pp 10132","article-title":"Auditing privacy defenses in federated learning via generative gradient leakage","author":"Li","year":"2022"},{"key":"mlstad4768bib3","first-page":"pp 9311","article-title":"Soteria: provable defense against privacy leakage in federated learning from representation perspective","author":"Sun","year":"2021"},{"article-title":"Understanding clipping for federated learning: Convergence and client-level differential privacy","year":"2022","author":"Zhang","key":"mlstad4768bib4"},{"key":"mlstad4768bib5","first-page":"pp 23296","article-title":"Intriguing properties of vision transformers","volume":"vol 34","author":"Naseer","year":"2021"},{"key":"mlstad4768bib6","doi-asserted-by":"publisher","first-page":"59","DOI":"10.1016\/j.ijmedinf.2018.01.007","article-title":"Federated learning of predictive models from federated electronic health records","volume":"112","author":"Brisimi","year":"2018","journal-title":"Int. J. Med. Inform."},{"key":"mlstad4768bib7","doi-asserted-by":"publisher","first-page":"945","DOI":"10.1093\/jamia\/ocy017","article-title":"Distributed deep learning networks among institutions for medical imaging","volume":"25","author":"Chang","year":"2018","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"mlstad4768bib8","first-page":"pp 2423","article-title":"Multi-institutional collaborations for improving deep learning-based magnetic resonance image reconstruction using federated learning","author":"Guo","year":"2021"},{"key":"mlstad4768bib9","first-page":"pp 1013","article-title":"Feddg: Federated domain generalization on medical image segmentation via episodic learning in continuous frequency space","author":"Liu","year":"2021"},{"key":"mlstad4768bib10","first-page":"pp 167","article-title":"Federated learning for data and model heterogeneity in medical imaging","author":"Madni","year":"2023"},{"key":"mlstad4768bib11","doi-asserted-by":"publisher","first-page":"16549","DOI":"10.1109\/ACCESS.2023.3246126","article-title":"Blockchain-based swarm learning for the mitigation of gradient leakage in federated learning","volume":"11","author":"Madni","year":"2023","journal-title":"IEEE Access"},{"key":"mlstad4768bib12","doi-asserted-by":"publisher","DOI":"10.1142\/S0129065723500338","article-title":"Swarm-fhe: fully homomorphic encryption-based swarm learning for malicious clients","volume":"33","author":"Madni","year":"2023","journal-title":"Int. J. Neural Syst."},{"key":"mlstad4768bib13","doi-asserted-by":"publisher","first-page":"6230","DOI":"10.3390\/s20216230","article-title":"Federated learning in smart city sensing: Challenges and opportunities","volume":"20","author":"Jiang","year":"2020","journal-title":"Sensors"},{"key":"mlstad4768bib14","doi-asserted-by":"publisher","first-page":"249","DOI":"10.1016\/j.neunet.2018.07.011","article-title":"A systematic study of the class imbalance problem in convolutional neural networks","volume":"106","author":"Buda","year":"2018","journal-title":"Neural Netw."},{"key":"mlstad4768bib15","article-title":"Learning imbalanced datasets with label-distribution-aware margin loss","volume":"vol 32","author":"Cao","year":"2019"},{"key":"mlstad4768bib16","first-page":"pp 19290","article-title":"Rethinking the value of labels for improving class-imbalanced learning","volume":"vol 33","author":"Yang","year":"2020"},{"key":"mlstad4768bib17","first-page":"pp 9268","article-title":"Class-balanced loss based on effective number of samples","author":"Cui","year":"2019"},{"key":"mlstad4768bib18","first-page":"pp 2537","article-title":"Large-scale long-tailed recognition in an open world","author":"Liu","year":"2019"},{"key":"mlstad4768bib19","first-page":"pp 4175","article-title":"Balanced meta-softmax for long-tailed visual recognition","volume":"vol 33","author":"Ren","year":"2020"},{"key":"mlstad4768bib20","first-page":"pp 11842","article-title":"Delving into deep imbalanced regression","author":"Yang","year":"2021"},{"key":"mlstad4768bib21","first-page":"pp 10713","article-title":"Model-contrastive federated learning","author":"Li","year":"2021"},{"key":"mlstad4768bib22","first-page":"pp 429","article-title":"Federated optimization in heterogeneous networks","volume":"vol 2","author":"Li","year":"2020"},{"key":"mlstad4768bib23","first-page":"pp 10112","article-title":"Feddc: Federated learning with non-iid data via local drift decoupling and correction","author":"Gao","year":"2022"},{"key":"mlstad4768bib24","first-page":"pp 16312","article-title":"Rethinking federated learning with domain shift: a prototype view","author":"Huang","year":"2023"},{"key":"mlstad4768bib25","doi-asserted-by":"crossref","first-page":"3363","DOI":"10.1109\/TIFS.2023.3279587","article-title":"Multi-domain virtual network embedding algorithm based on horizontal federated learning","volume":"18","author":"Zhang","year":"2023","journal-title":"IEEE Trans. Inform. Forensics Secur."},{"key":"mlstad4768bib26","article-title":"Attention is all you need","volume":"vol 30","author":"Vaswani","year":"2017"},{"key":"mlstad4768bib27","first-page":"pp 1","article-title":"An image is worth 16\u00d716 words: transformers for image recognition at scale","author":"Dosovitskiy","year":"2021"},{"key":"mlstad4768bib28","doi-asserted-by":"publisher","first-page":"700","DOI":"10.1093\/jamia\/ocaa017","article-title":"Accounting for data variability in multi-institutional distributed deep learning for medical imaging","volume":"27","author":"Balachandar","year":"2020","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"mlstad4768bib29","first-page":"pp 10012","article-title":"Swin transformer: hierarchical vision transformer using shifted windows","author":"Liu","year":"2021"},{"year":"2009","author":"Krizhevsky","key":"mlstad4768bib30"},{"key":"mlstad4768bib31","first-page":"pp 3730","article-title":"Deep learning face attributes in the wild","author":"Liu","year":"2015"},{"key":"mlstad4768bib32","first-page":"pp 10231","article-title":"Understanding robustness of transformers for image classification","author":"Bhojanapalli","year":"2021"},{"key":"mlstad4768bib33","first-page":"p 6558","article-title":"Multimodal transformer for unaligned multimodal language sequences","volume":"vol 2019","author":"Tsai","year":"2019"},{"article-title":"Bert: pre-training of deep bidirectional transformers for language understanding","year":"2018","author":"Devlin","key":"mlstad4768bib34"},{"key":"mlstad4768bib35","first-page":"pp 1439","article-title":"Unit: multimodal multitask learning with a unified transformer","author":"Hu","year":"2021"},{"key":"mlstad4768bib36","first-page":"pp 2071","article-title":"Vision transformers are robust learners","volume":"vol 36","author":"Paul","year":"2022"},{"key":"mlstad4768bib37","first-page":"pp 770","article-title":"Deep residual learning for image recognition","author":"He","year":"2016"},{"article-title":"Adaptive federated optimization","year":"2020","author":"Reddi","key":"mlstad4768bib38"},{"key":"mlstad4768bib39","first-page":"pp 67","article-title":"Image classification using lenet","author":"Verdhan","year":"2021"},{"article-title":"Deep long-tailed learning: a survey","year":"2021","author":"Zhang","key":"mlstad4768bib40"},{"key":"mlstad4768bib41","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1613\/jair.953","article-title":"Smote: synthetic minority over-sampling technique","volume":"16","author":"Chawla","year":"2002","journal-title":"J. Artif. Intell. Res."},{"key":"mlstad4768bib42","first-page":"pp 1322","article-title":"Adasyn: adaptive synthetic sampling approach for imbalanced learning","author":"He","year":"2008"},{"key":"mlstad4768bib43","doi-asserted-by":"publisher","first-page":"1367","DOI":"10.1109\/TPAMI.2018.2832629","article-title":"Imbalanced deep learning by minority class incremental rectification","volume":"41","author":"Dong","year":"2018","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"mlstad4768bib44","doi-asserted-by":"publisher","first-page":"2781","DOI":"10.1109\/TPAMI.2019.2914680","article-title":"Deep imbalanced learning for face recognition and attribute prediction","volume":"42","author":"Huang","year":"2019","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"mlstad4768bib45","first-page":"pp 6918","article-title":"Targeted supervised contrastive learning for long-tailed recognition","author":"Li","year":"2022"},{"article-title":"Long-tailed recognition by routing diverse distribution-aware experts","year":"2020","author":"Wang","key":"mlstad4768bib46"},{"article-title":"Self-supervised aggregation of diverse experts for test-agnostic long-tailed recognition","year":"2021","author":"Zhang","key":"mlstad4768bib47"},{"key":"mlstad4768bib48","article-title":"Meta-weight-net: learning an explicit mapping for sample weighting","volume":"vol 32","author":"Shu","year":"2019"},{"key":"mlstad4768bib49","first-page":"pp 5409","article-title":"Range loss for deep face recognition with long-tailed training data","author":"Zhang","year":"2017"},{"key":"mlstad4768bib50","doi-asserted-by":"publisher","first-page":"123","DOI":"10.1007\/s10994-009-5148-0","article-title":"Multi-domain learning by confidence-weighted parameter combination","volume":"79","author":"Dredze","year":"2010","journal-title":"Mach. Learn."},{"key":"mlstad4768bib51","doi-asserted-by":"publisher","first-page":"1345","DOI":"10.1109\/TKDE.2009.191","article-title":"A survey on transfer learning","volume":"22","author":"Pan","year":"2010","journal-title":"IEEE Trans. Knowl. Data Eng."},{"article-title":"Multi-domain adversarial learning","year":"2019","author":"Schoenauer-Sebag","key":"mlstad4768bib52"},{"key":"mlstad4768bib53","first-page":"pp 1249","article-title":"Learning deep feature representations with domain guided dropout for person re-identification","author":"Xiao","year":"2016"},{"key":"mlstad4768bib54","first-page":"2096","article-title":"Domain-adversarial training of neural networks","volume":"17","author":"Ganin","year":"2016","journal-title":"J. Mach. Learn. Res."},{"key":"mlstad4768bib55","first-page":"pp 1","article-title":"Domain generalization via conditional invariant representations","volume":"vol 32","author":"Li","year":"2018"},{"key":"mlstad4768bib56","first-page":"pp 443","article-title":"Deep coral: correlation alignment for deep domain adaptation","volume":"vol 14","author":"Sun","year":"2016"},{"article-title":"A unified perspective on multi-domain and multi-task learning","year":"2014","author":"Yang","key":"mlstad4768bib57"},{"key":"mlstad4768bib58","first-page":"pp 9719","article-title":"Bbn: bilateral-branch network with cumulative learning for long-tailed visual recognition","author":"Zhou","year":"2020"},{"key":"mlstad4768bib59","first-page":"pp 2229","article-title":"Domain generalization by solving jigsaw puzzles","author":"Carlucci","year":"2019"},{"article-title":"Domain generalization with mixstyle","year":"2021","author":"Zhou","key":"mlstad4768bib60"},{"key":"mlstad4768bib61","first-page":"pp 10","article-title":"Domain generalization via invariant feature representation","author":"Muandet","year":"2013"},{"key":"mlstad4768bib62","first-page":"pp 1","article-title":"Learning to generalize: Meta-learningfor domain generalization","volume":"vol 32","author":"Li","year":"2018"},{"key":"mlstad4768bib63","first-page":"pp 23664","article-title":"Adaptive risk minimization: learning to adapt to domain shift","volume":"vol 34","author":"Zhang","year":"2021"},{"article-title":"Invariant risk minimization","year":"2019","author":"Arjovsky","key":"mlstad4768bib64"},{"key":"mlstad4768bib65","first-page":"pp 5815","article-title":"Out-of-distribution generalization via risk extrapolation (rex)","author":"Krueger","year":"2021"},{"key":"mlstad4768bib66","first-page":"pp 10061","article-title":"Rethinking architecture design for tackling data heterogeneity in federated learning","author":"Qu","year":"2022"},{"article-title":"Split learning for health: distributed deep learning without sharing raw patient data","year":"2018","author":"Vepakomma","key":"mlstad4768bib67"},{"key":"mlstad4768bib68","doi-asserted-by":"crossref","DOI":"10.21203\/rs.3.rs-1087025\/v1","article-title":"Addressing catastrophic forgetting for medical domain expansion","author":"Gupta","year":"2021"},{"key":"mlstad4768bib69","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41598-020-69250-1","article-title":"Federated learning in medicine: facilitating multi-institutional collaborations without sharing patient data","volume":"10","author":"Sheller","year":"2020","journal-title":"Sci. Rep."},{"key":"mlstad4768bib70","first-page":"pp 4387","article-title":"The non-iid data quagmire of decentralized machine learning","author":"Hsieh","year":"2020"},{"key":"mlstad4768bib71","doi-asserted-by":"publisher","first-page":"4635","DOI":"10.1109\/JBHI.2022.3185956","article-title":"Splitavg: a heterogeneity-aware federated deep learning method for medical imaging","volume":"26","author":"Zhang","year":"2022","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"mlstad4768bib72","first-page":"pp 4912","article-title":"Collaborative unsupervised visual representation learning from decentralized data","author":"Zhuang","year":"2021"},{"article-title":"Measuring the effects of non-identical data distribution for federated visual classification","year":"2019","author":"Hsu","key":"mlstad4768bib73"},{"article-title":"Federated learning with matched averaging","year":"2020","author":"Wang","key":"mlstad4768bib74"},{"key":"mlstad4768bib75","first-page":"pp 4420","article-title":"Federated learning for non-iid data via unified feature learning and optimization objective alignment","author":"Zhang","year":"2021"},{"key":"mlstad4768bib76","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2022.102424","article-title":"Handling data heterogeneity with generative replay in collaborative learning for medical imaging","volume":"78","author":"Qu","year":"2022","journal-title":"Med. Image Anal."},{"key":"mlstad4768bib77","first-page":"pp 15076","article-title":"Ensemble attention distillation for privacy-preserving federated learning","author":"Gong","year":"2021"},{"key":"mlstad4768bib78","doi-asserted-by":"publisher","first-page":"3521","DOI":"10.1073\/pnas.1611835114","article-title":"Overcoming catastrophic forgetting in neural networks","volume":"114","author":"Kirkpatrick","year":"2017","journal-title":"Proc. Natl Acad. Sci."},{"key":"mlstad4768bib79","article-title":"Continual learning with deep generative replay","volume":"vol 30","author":"Shin","year":"2017"},{"key":"mlstad4768bib80","first-page":"pp 4055","article-title":"Image transformer","author":"Parmar","year":"2018"},{"article-title":"An image is worth 16x16 words: transformers for image recognition at scale","year":"2020","author":"Dosovitskiy","key":"mlstad4768bib81"},{"key":"mlstad4768bib82","doi-asserted-by":"publisher","first-page":"2278","DOI":"10.1109\/5.726791","article-title":"Gradient-based learning applied to document recognition","volume":"86","author":"LeCun","year":"1998","journal-title":"Proc. IEEE"},{"key":"mlstad4768bib83","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput."},{"article-title":"Pretrained transformers as universal computation engines","year":"2021","author":"Lu","key":"mlstad4768bib84"},{"key":"mlstad4768bib85","first-page":"pp 5542","article-title":"Deeper, broader and artier domain generalization","author":"Li","year":"2017"},{"article-title":"Federated learning with personalization layers","year":"2019","author":"Arivazhagan","key":"mlstad4768bib86"},{"key":"mlstad4768bib87","first-page":"pp 10122","article-title":"Differentially private federated learning with local regularization and sparsification","author":"Cheng","year":"2022"},{"key":"mlstad4768bib88","first-page":"pp 6105","article-title":"Efficientnet: rethinking model scaling for convolutional neural networks","author":"Tan","year":"2019"},{"key":"mlstad4768bib89","first-page":"pp 57","article-title":"On multi-domain long-tailed recognition, imbalanced domain generalization and beyond","author":"Yang","year":"2022"},{"article-title":"The mnist database of handwritten digits (2010)","year":"2009","author":"LeCun","key":"mlstad4768bib90"},{"article-title":"Reading digits in natural images with unsupervised feature learning","year":"2011","author":"Netzer","key":"mlstad4768bib91"},{"key":"mlstad4768bib92","doi-asserted-by":"publisher","first-page":"550","DOI":"10.1109\/34.291440","article-title":"A database for handwritten text recognition research","volume":"16","author":"Hull","year":"1994","journal-title":"IEEE Trans. Pattern Anal. Mach. Intell."},{"key":"mlstad4768bib93","first-page":"pp 437","article-title":"Auto-fedrl: federated hyperparameter optimization for multi-institutional medical image segmentation","author":"Guo","year":"2022"},{"article-title":"Adam: a method for stochastic optimization","year":"2014","author":"Kingma","key":"mlstad4768bib94"},{"article-title":"Federated learning with non-iid data","year":"2018","author":"Zhao","key":"mlstad4768bib95"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad4768","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad4768\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad4768","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad4768\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad4768\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad4768\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad4768\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad4768\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,5,16]],"date-time":"2024-05-16T07:56:25Z","timestamp":1715846185000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/ad4768"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,5,16]]},"references-count":95,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2024,5,16]]},"published-print":{"date-parts":[[2024,6,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/ad4768","relation":{},"ISSN":["2632-2153"],"issn-type":[{"type":"electronic","value":"2632-2153"}],"subject":[],"published":{"date-parts":[[2024,5,16]]},"assertion":[{"value":"Exploiting data diversity in multi-domain federated learning","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2024 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2023-11-19","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2024-05-03","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2024-05-16","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}