AI Paper Atlas

Research map / Federated learning and privacy / Personalized and heterogeneous FL

Robust Heterogeneous Federated Learning under Data Corruption

Xiuwen Fang, Mang Ye, Xiyuan Yang

ICCV 2023 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Non-IID federated learning · 50 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add federated communication (HFL) under data corruption (ablation)local models · CIFAR-10-C xi=0.5, corrupted testavg accuracy: 58.01 -> 56.08Hurt
add HFL on clean data (ablation)local models · CIFAR-10-C xi=0, clean testavg accuracy: 77.98 -> 78.61Improved
add PubAug re-weighted communication (ablation)HFL · CIFAR-10-C xi=0, corrupted testavg accuracy: 63.45 -> 64.88Improved
add PubAug when all clients fully corrupted (ablation)HFL · CIFAR-10-C xi=1, corrupted testavg accuracy: 57.71 -> 56.45No clear effect
add corruption-robust augmentation (Aug) without JSD (ablation)HFL + PubAug · CIFAR-10-C xi=0, clean testavg accuracy: 78.79 -> 72.10Hurt
add JSD consistency loss on top of Aug (ablation)AugHFL · CIFAR-10-C xi=0.5, corrupted testavg accuracy: 57.44 -> 68.52Improved
add JSD on top of Aug (ablation)AugHFL · CIFAR-10-C xi=0.5, clean testavg accuracy: 66.19 -> 73.17Improved
use AugHFL instead of prior heterogeneous FL methods when training data heavily corruptedFedMD/FedDF/RHFL/FCCL · CIFAR-10-C xi=1, corrupted testavg accuracy: 57.91 -> 70.03 vs FCCLImproved
use AugHFL in fully clean settingAugHFL vs FCCL · CIFAR-10-C xi=0, clean testavg accuracy: 80.39 -> 77.86Hurt
increase private-data corruption rate from 0 to 1AugHFL · CIFAR-10-Cavg accuracy: 4.2% on clean, 1.18% on corruptedHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy