Poster
in
Workshop: INTERPOLATE — First Workshop on Interpolation Regularizers and Beyond
FedLN: Federated Learning with Label Noise
Vasileios Tsouvalas · Aaqib Saeed · Tanir Özçelebi · Nirvana Meratnia
Keywords: [ Deep Learning ] [ label correction ] [ federated learning ] [ noisy labels ]
Federated Learning (FL) is a distributed machine learning paradigm that enables learning models from decentralized private datasets, where the labeling effort is entrusted to the clients. While most existing FL approaches assume high-quality labels are readily available on users' devices; in reality, label noise can naturally occur in FL and follows a non-i.i.d. distribution among clients. Due to the ``non-iid-ness'' challenges, existing state-of-the-art centralized approaches exhibit unsatisfactory performance, while previous FL studies rely on data exchange or repeated server-side aid to improve model's performance. Here, we propose FedLN, a framework to deal with label noise across different FL training stages; namely, FL initialization, and server-side model aggregation. Extensive experiments on various publicly available vision and audio datasets demonstrate an improvement of 24% on average compared to state-of-the-art methods for a label noise level of 70%.