Federated Learning with Bilateral Curation for Partially Class-Disjoint Data

TL;DR

This paper introduces FedGELA, a federated learning method that uses a fixed simplex ETF classifier to address challenges of partially class-disjoint data, improving global and local performance.

Contribution

The paper proposes FedGELA, a novel federated learning approach with a globally fixed simplex ETF classifier, effectively handling PCDD challenges and enhancing performance.

Findings

Achieves an average of 3.9% improvement over FedAvg

Provides both local and global convergence guarantees

Demonstrates promising results on multiple datasets

Abstract

Partially class-disjoint data (PCDD), a common yet under-explored data formation where each client contributes a part of classes (instead of all classes) of samples, severely challenges the performance of federated algorithms. Without full classes, the local objective will contradict the global objective, yielding the angle collapse problem for locally missing classes and the space waste problem for locally existing classes. As far as we know, none of the existing methods can intrinsically mitigate PCDD challenges to achieve holistic improvement in the bilateral views (both global view and local view) of federated learning. To address this dilemma, we are inspired by the strong generalization of simplex Equiangular Tight Frame~(ETF) on the imbalanced data, and propose a novel approach called FedGELA where the classifier is globally fixed as a simplex ETF while locally adapted to the personal distributions. Globally, FedGELA provides fair and equal discrimination for all classes and avoids inaccurate updates of the classifier, while locally it utilizes the space of locally missing classes for locally existing classes. We conduct extensive experiments on a range of datasets to demonstrate that our FedGELA achieves promising performance~(averaged improvement of 3.9% to FedAvg and 1.5% to best baselines) and provide both local and global convergence guarantees. Source code is available at:https://github.com/MediaBrain-SJTU/FedGELA.git.