5 Papers
Bin Wu is an academic researcher from Chinese Academy of Sciences. The author has contributed to research in topics: Computer science & Differential privacy. The author has an hindex of 1, co-authored 1 publications.
Chat about Author
Papers
Sensitive attribute privacy preservation of trajectory data publishing based on l-diversity
TL;DR: A data privacy preservation with enhanced l-diversity by determining those critical spatial-temporal sequences which are more likely to cause privacy leakage and perturbing these sequences to achieve better privacy while still ensuring high utility compared with existing privacy preservation schemes on trajectory.
Privacy Preservation for Trajectory Publication Based on Differential Privacy
TL;DR: This article proposes a comprehensive trajectory publishing algorithm with three effective procedures that effectively protects the privacy of both sensitive labels and location data in trajectory publication and can also achieve higher data utility.
15
DLDP-FL: Dynamic local differential privacy federated learning method based on mesh network edge devices
TL;DR: Wang et al. as discussed by the authors proposed a novel edge FL architecture based on edge devices in mesh network architecture; next, they exploited the mesh networking features to address the problem of possible internal attacks from edge devices and design a Dynamic Local Differential Privacy (DLDP) algorithm; then, according to the communication characteristics of mesh network, they design Edge-FedAvg algorithm to reduce the communication cost; finally, to enhance the response to untrusted center servers, embed watermark in the model to further enhance the privacy protection capability.
3
NBA: defensive distillation for backdoor removal via neural behavior alignment
Zonghao Ying,Bin Wu +1 more
TL;DR: In this paper , the authors propose Neural Behavioral Alignment (NBA) to align the backdoor neural behavior from the student network with the benign neural behaviour from the teacher network, which enables the proactive removal of backdoors.
DLP: towards active defense against backdoor attacks with decoupled learning process
Zonghao Ying,Bin Wu +1 more
TL;DR: In this article , the authors propose a general training pipeline to defend against backdoor attacks by decoupling the learning process into three stages, i.e., supervised learning, active unlearning, and active semi-supervised fine-tuning.