Vertical Federated Learning: Concepts, Advances and Challenges

  • 2023-09-28 02:29:46
  • Yang Liu, Yan Kang, Tianyuan Zou, Yanhong Pu, Yuanqin He, Xiaozhou Ye, Ye Ouyang, Ya-Qin Zhang, Qiang Yang
  • 0

Abstract

Vertical Federated Learning (VFL) is a federated learning setting wheremultiple parties with different features about the same set of users jointlytrain machine learning models without exposing their raw data or modelparameters. Motivated by the rapid growth in VFL research and real-worldapplications, we provide a comprehensive review of the concept and algorithmsof VFL, as well as current advances and challenges in various aspects,including effectiveness, efficiency, and privacy. We provide an exhaustivecategorization for VFL settings and privacy-preserving protocols andcomprehensively analyze the privacy attacks and defense strategies for eachprotocol. In the end, we propose a unified framework, termed VFLow, whichconsiders the VFL problem under communication, computation, privacy, as well aseffectiveness and fairness constraints. Finally, we review the most recentadvances in industrial applications, highlighting open challenges and futuredirections for VFL.

 

Quick Read (beta)

loading the full paper ...