Learning Cautiously in Federated Learning with Noisy and Heterogeneous Clients

2023/04/06 by Wu, Chenrui, Li, Zexi, Wang, Fangxin +1 · 2 citations
#Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.2304.02892

Abstract

Federated learning (FL) is a distributed framework for collaboratively training with privacy guarantees. In real-world scenarios, clients may have Non-IID data (local class imbalance) with poor annotation quality (label noise). The co-existence of label noise and class imbalance in FL's small local datasets renders conventional FL methods and noisy-label learning methods both ineffective. To address the challenges, we propose FedCNI without using an additional clean proxy dataset. It includes a noise-resilient local solver and a robust global aggregator. For the local solver, we design a more robust prototypical noise detector to distinguish noisy samples. Further to reduce the negative impact brought by the noisy samples, we devise a curriculum pseudo labeling method and a denoise Mixup training strategy. For the global aggregator, we propose a switching re-weighted aggregation method tailored to different learning periods. Extensive experiments demonstrate our method can substantially outperform state-of-the-art solutions in mix-heterogeneous FL environments.

Cited by

• Robust Federated Learning with Confidence-Weighted Filtering and GAN-Based Completion under Noisy and Incomplete Data
• FNBench: Benchmarking Robust Federated Learning against Noisy Labels

Related

• FedEFC: Federated Learning Using Enhanced Forward Correction Against Noisy Labels
• FedNoiL: A Simple Two-Level Sampling Method for Federated Learning with Noisy Labels
• FedNoRo: Towards Noise-Robust Federated Learning by Addressing Class Imbalance and Label Noise Heterogeneity
• Federated Noisy Client Learning
• FedDiv: Collaborative Noise Filtering for Federated Learning with Noisy Labels