Show simple item record

dc.contributor.authorHan, Yufei
dc.contributor.authorZhang, Xiangliang
dc.date.accessioned2020-09-20T11:20:44Z
dc.date.available2020-09-20T11:20:44Z
dc.date.issued2020-04-03
dc.identifier.citationHan, Y., & Zhang, X. (2020). Robust Federated Learning via Collaborative Machine Teaching. Proceedings of the AAAI Conference on Artificial Intelligence, 34(04), 4075–4082. doi:10.1609/aaai.v34i04.5826
dc.identifier.issn2374-3468
dc.identifier.issn2159-5399
dc.identifier.doi10.1609/aaai.v34i04.5826
dc.identifier.urihttp://hdl.handle.net/10754/665233
dc.description.abstractFor federated learning systems deployed in the wild, data flaws hosted on local agents are widely witnessed. On one hand, given a large amount (e.g. over 60%) of training data are corrupted by systematic sensor noise and environmental perturbations, the performances of federated model training can be degraded significantly. On the other hand, it is prohibitively expensive for either clients or service providers to set up manual sanitary checks to verify the quality of data instances. In our study, we echo this challenge by proposing a collaborative and privacy-preserving machine teaching method. Specifically, we use a few trusted instances provided by teachers as benign examples in the teaching process. Our collaborative teaching approach seeks jointly the optimal tuning on the distributed training set, such that the model learned from the tuned training set predicts labels of the trusted items correctly. The proposed method couples the process of teaching and learning and thus produces directly a robust prediction model despite the extremely pervasive systematic data corruption. The experimental study on real benchmark data sets demonstrates the validity of our method.
dc.publisherAssociation for the Advancement of Artificial Intelligence (AAAI)
dc.relation.urlhttps://aaai.org/ojs/index.php/AAAI/article/view/5826
dc.rightsArchived with thanks to Proceedings of the AAAI Conference on Artificial Intelligence
dc.titleRobust Federated Learning via Collaborative Machine Teaching
dc.typeArticle
dc.contributor.departmentComputer Science Program
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.contributor.departmentMachine Intelligence & kNowledge Engineering Lab
dc.identifier.journalProceedings of the AAAI Conference on Artificial Intelligence
dc.eprint.versionPost-print
dc.contributor.institutionNortonlifelock Research Group Campus SohpiaTech Sophia Antipolis, 06410, France.
dc.identifier.volume34
dc.identifier.issue04
dc.identifier.pages4075-4082
kaust.personZhang, Xiangliang
refterms.dateFOA2020-09-21T05:47:57Z


Files in this item

Thumbnail
Name:
AAAI_submission.pdf
Size:
274.9Kb
Format:
PDF
Description:
Accepted manuscript

This item appears in the following Collection(s)

Show simple item record