Please use this identifier to cite or link to this item: https://elib.vku.udn.vn/handle/123456789/3990
Full metadata record
DC FieldValueLanguage
dc.contributor.authorLe, Huy Q.-
dc.contributor.authorNguyen, Huu Nhat Minh-
dc.contributor.authorPandey, Shashi Raj-
dc.contributor.authorZhang, Chaoning-
dc.contributor.authorHong, Choong Seon-
dc.date.accessioned2024-07-29T08:41:05Z-
dc.date.available2024-07-29T08:41:05Z-
dc.date.issued2024-07-
dc.identifier.urihttps://doi.org/10.1016/j.engappai.2024.108093-
dc.identifier.urihttps://elib.vku.udn.vn/handle/123456789/3990-
dc.descriptionEngineering Applications of Artificial Intelligence; Volume 133, Part B; pp: 108093vi_VN
dc.description.abstractIn a practical setting, how to enable robust Federated Learning (FL) systems, both in terms of generalization and personalization abilities, is one important research question. It is a challenging issue due to the consequences of non-i.i.d. properties of client’s data, often referred to as statistical heterogeneity, and small local data samples from the various data distributions. Therefore, to develop robust generalized global and personalized models, conventional FL methods need to redesign the knowledge aggregation from biased local models while considering huge divergence of learning parameters due to skewed client data. In this work, we demonstrate that the knowledge transfer mechanism achieves these objectives and develop a novel knowledge distillation-based approach to study the extent of knowledge transfer between the global model and local models. Henceforth, our method considers the suitability of transferring the outcome distribution and (or) the embedding vector of representation from trained models during cross-device knowledge transfer using a small proxy dataset in heterogeneous FL. In doing so, we alternatively perform cross-device knowledge transfer following general formulations as (1) global knowledge transfer and (2) on-device knowledge transfer. Through simulations on three federated datasets, we show the proposed method achieves significant speedups and high personalized performance of local models. Furthermore, the proposed approach offers a more stable algorithm than other baselines during the training, with minimal communication data load when exchanging the trained model’s outcomes and representation.vi_VN
dc.language.isoenvi_VN
dc.publisherElsevier Ltdvi_VN
dc.subjectFederated Learningvi_VN
dc.subjectdata distributionsvi_VN
dc.subjecttransferringvi_VN
dc.titleCDKT-FL: Cross-device knowledge transfer using proxy dataset in federated learningvi_VN
dc.typeWorking Papervi_VN
Appears in Collections:NĂM 2024

Files in This Item:

 Sign in to read



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.