A deep cut into Split Federated Self-supervised Learning

TL;DR

This paper introduces MonAcoSFL, a novel split federated self-supervised learning method that improves privacy and communication efficiency by aligning models, achieving state-of-the-art accuracy in distributed training.

Contribution

We propose MonAcoSFL, a new approach that addresses privacy and communication challenges in split federated self-supervised learning by model alignment.

Findings

MonAcoSFL outperforms existing methods in accuracy.

It significantly reduces communication overhead.

Splitting depth is crucial for privacy and efficiency.

Abstract

Collaborative self-supervised learning has recently become feasible in highly distributed environments by dividing the network layers between client devices and a central server. However, state-of-the-art methods, such as MocoSFL, are optimized for network division at the initial layers, which decreases the protection of the client data and increases communication overhead. In this paper, we demonstrate that splitting depth is crucial for maintaining privacy and communication efficiency in distributed training. We also show that MocoSFL suffers from a catastrophic quality deterioration for the minimal communication overhead. As a remedy, we introduce Momentum-Aligned contrastive Split Federated Learning (MonAcoSFL), which aligns online and momentum client models during training procedure. Consequently, we achieve state-of-the-art accuracy while significantly reducing the communication overhead, making MonAcoSFL more practical in real-world scenarios.