Y
Yang Kang
Publications - 12
Citations - 79
Yang Kang is an academic researcher. The author has contributed to research in topics: Computer science & Homomorphic encryption. The author has an hindex of 1, co-authored 1 publications receiving 1 citations.
Papers
More filters
Posted Content
Secure and Efficient Federated Transfer Learning
TL;DR: This work aims towards enhancing the efficiency and security of existing models for practical collaborative training under a data federation by incorporating Secret Sharing (SS), and improves upon the previous solution, and allows malicious players who can arbitrarily deviate from the protocol in the FTL model.
Journal ArticleDOI
Defending Batch-Level Label Inference and Replacement Attacks in Vertical Federated Learning
TL;DR: In this article , a Confusional AutoEncoder (CAE) based on autoencoder and entropy regularization was proposed to defend against gradient inversion attacks. But the authors also showed that even with HE-protected communication, private labels can still be reconstructed with high accuracy by gradient-inversion attack.
Journal ArticleDOI
Trading Off Privacy, Utility and Efficiency in Federated Learning
TL;DR: It is indicated that it is unrealistic to expect an FL algorithm to simultaneously provide excellent privacy, utility, and efficiency in certain scenarios, and a framework is proposed that reconciles horizontal and vertical federated learning.
Journal ArticleDOI
Vertical Federated Learning
Yang Liu,Yang Kang,Tianyuan Zou,Yanhong Pu,Yuanqin He,Xiaozhou Ye,Ye Ouyang,Yaqin Zhang,Qian Yang +8 more
TL;DR: In this paper , the authors provide a comprehensive review of the concept and algorithms of VFL, as well as current advances and challenges in various aspects, including effectiveness, efficiency, and privacy.
Journal ArticleDOI
A Framework for Evaluating Privacy-Utility Trade-off in Vertical Federated Learning
TL;DR: This work proposes an evaluation framework that formulates the privacy-utility evaluation problem and uses this framework as a guide to comprehensively evaluate a broad range of protection mechanisms against most of the state-of-the-art privacy attacks for three widely-deployed VFL algorithms.