… Federatedlearning (FL) is a kind of distributed machine learning framework, where the global model is generated on the centralized aggregation server based on the parameters of …
MR Sprague, A Jalalirad, M Scavuzzo… - … on Machine Learning …, 2018 - Springer
… We propose an asynchronous aggregation scheme for federatedlearning. We compare its performance to the baseline synchronous federated-averaging algorithm [7]. We show its …
… • We propose FedBuff, a novel asynchronousfederated optimization framework with buffered asynchronous aggregation to achieve scalability and privacy against the honest-but-…
Y Chen, Y Ning, M Slawski… - 2020 IEEE International …, 2020 - ieeexplore.ieee.org
… develop an asynchronous online federatedlearning framework with … We propose an asynchronous online federatedlearning … from lagging clients, and clients perform online learning …
Z Wang, Z Zhang, Y Tian, Q Yang… - IEEE Transactions …, 2022 - ieeexplore.ieee.org
… asynchronous FL framework, which well adapts to the heterogeneity of users, communication environments and learning … validate that the proposed asynchronous FL framework can …
X Lu, Y Liao, P Lio, P Hui - Ieee Access, 2020 - ieeexplore.ieee.org
… The state-of-art solutions of federatedlearning are … gradient compression and asynchronous federatedlearning with dual… correction to solve asynchronousfederatedlearning problems. …
J Liu, H Xu, L Wang, Y Xu, C Qian… - IEEE Transactions …, 2021 - ieeexplore.ieee.org
… The asynchronous scheme is proposed for FL. In the … we propose an adaptive asynchronous federatedlearning (AAFL) … We design an adaptive asynchronousfederatedlearning (AAFL) …
Q Ma, Y Xu, H Xu, Z Jiang, L Huang… - IEEE Journal on …, 2021 - ieeexplore.ieee.org
… asynchronous updating may seriously hurt training accuracy, especially on Non-IID data. In this paper, we propose a semi-asynchronousfederatedlearning … a federatedlearning system …
Y Lu, X Huang, Y Dai, S Maharjan… - IEEE Transactions on …, 2019 - ieeexplore.ieee.org
… models in federatedlearning by incorporating local differential privacy (LDP) into gradient descent training scheme. 2) We develop a new asynchronousfederatedlearning architecture …