Survey of Privacy Threats and Countermeasures in Federated Learning

TL;DR

This survey comprehensively categorizes and describes privacy threats and countermeasures across different types of federated learning, highlighting the unique challenges and solutions for horizontal, vertical, and transfer federated learning.

Contribution

It provides a detailed classification and analysis of privacy threats and countermeasures specific to various federated learning paradigms, which was lacking in prior work.

Findings

Identified common and unique privacy threats in federated learning types.

Reviewed existing countermeasures tailored to each federated learning paradigm.

Highlighted gaps and future directions for privacy protection in federated learning.

Abstract

Federated learning is widely considered to be as a privacy-aware learning method because no training data is exchanged directly between clients. Nevertheless, there are threats to privacy in federated learning, and privacy countermeasures have been studied. However, we note that common and unique privacy threats among typical types of federated learning have not been categorized and described in a comprehensive and specific way. In this paper, we describe privacy threats and countermeasures for the typical types of federated learning; horizontal federated learning, vertical federated learning, and transfer federated learning.