no code implementations • 23 Mar 2022 • Zexi Li, Jiaxun Lu, Shuang Luo, Didi Zhu, Yunfeng Shao, Yinchuan Li, Zhimeng Zhang, Yongheng Wang, Chao Wu
In the literature, centralized clustered FL algorithms require the assumption of the number of clusters and hence are not effective enough to explore the latent relationships among clients.
no code implementations • 2 Dec 2021 • Shuo Wan, Jiaxun Lu, Pingyi Fan, Yunfeng Shao, Chenghui Peng, Khaled B. Letaief
In this paper, we develop a vertical-horizontal federated learning (VHFL) process, where the global feature is shared with the agents in a procedure similar to that of vertical FL without any extra communication rounds.
no code implementations • 9 Nov 2021 • Fengda Zhang, Kun Kuang, Yuxuan Liu, Long Chen, Chao Wu, Fei Wu, Jiaxun Lu, Yunfeng Shao, Jun Xiao
We validate the advantages of the FMDA-M algorithm with various kinds of distribution shift settings in experiments, and the results show that FMDA-M algorithm outperforms the existing fair FL algorithms on unified group fairness.
no code implementations • 30 Apr 2021 • Shuo Wan, Jiaxun Lu, Pingyi Fan, Yunfeng Shao, Chenghui Peng, Khaled B. Letaief
Federated learning (FL) has recently emerged as an important and promising learning scheme in IoT, enabling devices to jointly learn a model without sharing their raw data sets.
no code implementations • 5 May 2019 • Shuo Wan, Jiaxun Lu, Pingyi Fan, Khaled B. Letaief
In this paper, a MEC-based big data analysis network is discussed.
no code implementations • 19 Nov 2018 • Yuheng Bu, Jiaxun Lu, Venugopal V. Veeravalli
The goal is to detect whether the change in the model is significant, i. e., whether the difference between the pre-change parameter and the post-change parameter $\|\theta-\theta'\|_2$ is larger than a pre-determined threshold $\rho$.
no code implementations • 29 May 2018 • Yuheng Bu, Jiaxun Lu, Venugopal V. Veeravalli
Furthermore, an estimator of the change in the learning problems using the active learning samples is constructed, which provides an adaptive sample size selection rule that guarantees the excess risk is bounded for sufficiently large number of time steps.