Prototype Guided Federated Learning of Visual Feature Representations

TL;DR

This paper introduces FedProto, a novel federated learning method that leverages prototypical feature representations and an attention mechanism to improve model accuracy, convergence, and uncertainty reduction in vision tasks.

Contribution

FedProto is the first approach to incorporate feature representation margins into federated learning, enhancing training effectiveness for vision tasks including dense prediction.

Findings

Achieves state-of-the-art accuracy and convergence in image classification and segmentation.

Reduces prediction uncertainty compared to baseline methods.

First to evaluate federated learning in dense prediction tasks.

Abstract

Federated Learning (FL) is a framework which enables distributed model training using a large corpus of decentralized training data. Existing methods aggregate models disregarding their internal representations, which are crucial for training models in vision tasks. System and statistical heterogeneity (e.g., highly imbalanced and non-i.i.d. data) further harm model training. To this end, we introduce a method, called FedProto, which computes client deviations using margins of prototypical representations learned on distributed data, and applies them to drive federated optimization via an attention mechanism. In addition, we propose three methods to analyse statistical properties of feature representations learned in FL, in order to elucidate the relationship between accuracy, margins and feature discrepancy of FL models. In experimental analyses, FedProto demonstrates state-of-the-art accuracy and convergence rate across image classification and semantic segmentation benchmarks by enabling maximum margin training of FL models. Moreover, FedProto reduces uncertainty of predictions of FL models compared to the baseline. To our knowledge, this is the first work evaluating FL models in dense prediction tasks, such as semantic segmentation.