ABC-FL: Anomalous and Benign client Classification in Federated Learning
This addresses data poisoning vulnerabilities in federated learning for privacy-preserving ML applications, but it is incremental as it builds on prior detection methods by handling non-IID benign data.
The paper tackles the problem of detecting malicious clients in federated learning when benign clients have non-IID data, proposing a method that uses feature dimension reduction, dynamic clustering, and cosine similarity-based clipping, and it shows the method classifies malicious clients and reduces their negative impact.
Federated Learning is a distributed machine learning framework designed for data privacy preservation i.e., local data remain private throughout the entire training and testing procedure. Federated Learning is gaining popularity because it allows one to use machine learning techniques while preserving privacy. However, it inherits the vulnerabilities and susceptibilities raised in deep learning techniques. For instance, Federated Learning is particularly vulnerable to data poisoning attacks that may deteriorate its performance and integrity due to its distributed nature and inaccessibility to the raw data. In addition, it is extremely difficult to correctly identify malicious clients due to the non-Independently and/or Identically Distributed (non-IID) data. The real-world data can be complex and diverse, making them hardly distinguishable from the malicious data without direct access to the raw data. Prior research has focused on detecting malicious clients while treating only the clients having IID data as benign. In this study, we propose a method that detects and classifies anomalous clients from benign clients when benign ones have non-IID data. Our proposed method leverages feature dimension reduction, dynamic clustering, and cosine similarity-based clipping. The experimental results validates that our proposed method not only classifies the malicious clients but also alleviates their negative influences from the entire procedure. Our findings may be used in future studies to effectively eliminate anomalous clients when building a model with diverse data.