FedX: Unsupervised Federated Learning with Cross Knowledge Distillation
This addresses the challenge of privacy-preserving unsupervised learning in federated settings for applications like distributed data analysis, though it appears incremental as an add-on module.
The paper tackles the problem of learning unbiased representations from decentralized and heterogeneous local data in unsupervised federated learning, resulting in performance improvements of 1.58 to 5.52 percentage points on five unsupervised algorithms.
This paper presents FedX, an unsupervised federated learning framework. Our model learns unbiased representation from decentralized and heterogeneous local data. It employs a two-sided knowledge distillation with contrastive learning as a core component, allowing the federated system to function without requiring clients to share any data features. Furthermore, its adaptable architecture can be used as an add-on module for existing unsupervised algorithms in federated settings. Experiments show that our model improves performance significantly (1.58--5.52pp) on five unsupervised algorithms.