0

L-DAWA: Layer-wise Divergence Aware Weight Aggregation in Federated Self-Supervised Visual Representation Learning

A new layer-wise aggregation strategy mitigates client bias and divergence in federated learning, improving visual representation quality and robustness on unlabeled edge data.

Year
2023
Venue
ICCV 2023 1
Authors
6
Hosting
Abstract onlyARXIV-DEFAULT

Cite

Notes

Only stored in your browser.

Attribution

Abstract & full text
arxiv.org/abs/2307.07393ARXIV-DEFAULT
TL;DR
Semantic Scholar
Attribution policy →

Abstract

The ubiquity of camera-enabled devices has led to large amounts of unlabeled image data being produced at the edge. The integration of self-supervised learning (SSL) and federated learning (FL) into one coherent system can potentially offer data privacy guarantees while also advancing the quality and robustness of the learned visual representations without needing to move data around. However, client bias and divergence during FL aggregation caused by data heterogeneity limits the performance of learned visual representations on downstream tasks. In this paper, we propose a new aggregation strategy termed Layer-wise Divergence Aware Weight Aggregation (L-DAWA) to mitigate the influence of client bias and divergence during FL aggregation. The proposed method aggregates weights at the layer-level according to the measure of angular divergence between the clients' model and the global model. Extensive experiments with cross-silo and cross-device settings on CIFAR-10/100 and Tiny ImageNet datasets demonstrate that our methods are effective and obtain new SOTA performance on both contrastive and non-contrastive SSL approaches.

Authors

6