SemiFL: Semi-Supervised Federated Learning for Unlabeled Clients with Alternate Training

TL;DR

SemiFL introduces an innovative semi-supervised federated learning framework that effectively trains models with unlabeled client data by alternating between global model fine-tuning and pseudo-label generation, improving performance and communication efficiency.

Contribution

The paper proposes a novel alternate training method for semi-supervised federated learning, addressing the challenge of unlabeled client data and enhancing model accuracy with minimal communication.

Findings

Significant performance improvement over existing SSFL baselines

Outperforms many state-of-the-art FL and SSL methods

Effective use of unlabeled data with alternate training strategy

Abstract

Federated Learning allows the training of machine learning models by using the computation and private data resources of many distributed clients. Most existing results on Federated Learning (FL) assume the clients have ground-truth labels. However, in many practical scenarios, clients may be unable to label task-specific data due to a lack of expertise or resource. We propose SemiFL to address the problem of combining communication-efficient FL such as FedAvg with Semi-Supervised Learning (SSL). In SemiFL, clients have completely unlabeled data and can train multiple local epochs to reduce communication costs, while the server has a small amount of labeled data. We provide a theoretical understanding of the success of data augmentation-based SSL methods to illustrate the bottleneck of a vanilla combination of communication-efficient FL with SSL. To address this issue, we propose alternate training to `fine-tune global model with labeled data' and `generate pseudo-labels with the global model.' We conduct extensive experiments and demonstrate that our approach significantly improves the performance of a labeled server with unlabeled clients training with multiple local epochs. Moreover, our method outperforms many existing SSFL baselines and performs competitively with the state-of-the-art FL and SSL results.