Feature Distribution Matching for Federated Domain Generalization

TL;DR

This paper introduces FedKA, a federated domain generalization method that aligns feature distributions across clients to improve model robustness on unseen tasks without sharing raw data.

Contribution

FedKA is a novel federated learning approach that matches feature distributions and uses a voting mechanism for pseudo-labeling, enhancing domain invariance and reducing negative transfer.

Findings

FedKA improves accuracy by 8.8% on Digit-Five.

FedKA achieves a 3.5% gain on Office-Caltech10.

FedKA reduces negative transfer, quadrupling performance gains.

Abstract

Multi-source domain adaptation has been intensively studied. The distribution shift in features inherent to specific domains causes the negative transfer problem, degrading a model's generality to unseen tasks. In Federated Learning (FL), learned model parameters are shared to train a global model that leverages the underlying knowledge across client models trained on separate data domains. Nonetheless, the data confidentiality of FL hinders the effectiveness of traditional domain adaptation methods that require prior knowledge of different domain data. We propose a new federated domain generalization method called Federated Knowledge Alignment (FedKA). FedKA leverages feature distribution matching in a global workspace such that the global model can learn domain-invariant client features under the constraint of unknown client data. FedKA employs a federated voting mechanism that generates target domain pseudo-labels based on the consensus from clients to facilitate global model fine-tuning. We performed extensive experiments, including an ablation study, to evaluate the effectiveness of the proposed method in both image and text classification tasks using different model architectures. The empirical results show that FedKA achieves performance gains of 8.8% and 3.5% in Digit-Five and Office-Caltech10, respectively, and a gain of 0.7% in Amazon Review with extremely limited training data. Moreover, we studied the effectiveness of FedKA in alleviating the negative transfer of FL based on a new criterion called Group Effect. The results show that FedKA can reduce negative transfer, improving the performance gain via model aggregation by 4 times.