Revisiting Early-Learning Regularization When Federated Learning Meets Noisy Labels
It addresses label noise in federated learning, a domain-specific challenge for decentralized and heterogeneous data, but is incremental as it builds on existing early-learning regularization and FL techniques.
This paper tackles the problem of label noise in federated learning by introducing Federated Label-mixture Regularization (FLR), which blends local and global model predictions to generate pseudo labels, resulting in enhanced accuracy and reduced memorization of noisy labels in both i.i.d. and non-i.i.d. settings.
In the evolving landscape of federated learning (FL), addressing label noise presents unique challenges due to the decentralized and diverse nature of data collection across clients. Traditional centralized learning approaches to mitigate label noise are constrained in FL by privacy concerns and the heterogeneity of client data. This paper revisits early-learning regularization, introducing an innovative strategy, Federated Label-mixture Regularization (FLR). FLR adeptly adapts to FL's complexities by generating new pseudo labels, blending local and global model predictions. This method not only enhances the accuracy of the global model in both i.i.d. and non-i.i.d. settings but also effectively counters the memorization of noisy labels. Demonstrating compatibility with existing label noise and FL techniques, FLR paves the way for improved generalization in FL environments fraught with label inaccuracies.