AI Paper Atlas

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

FedCorr: Multi-Stage Federated Learning for Label Noise Correction

Jingyi Xu, Zihan Chen, Tony Q.S. Quek, Kai Fong Ernest Chong

CVPR 2022 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated learning with pretrained models · 121 citations (OpenAlex)

Check 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
Apply FedCorr's multi-stage identification and label correction frameworkfederated training · CIFAR-10 IID, (ρ,τ)=(0.8,0.5)best test accuracy (%): 80.59±1.49 -> 90.59±0.70Improved
Apply FedCorr's multi-stage identification and label correction frameworkfederated training · CIFAR-100 IID, (ρ,τ)=(0.8,0.5)best test accuracy (%): 51.24±1.60 -> 59.10±5.12Improved
Use the proposed non-IID client partition with FedCorrCIFAR-10 federated learning · CIFAR-10 non-IID, (p,αDir)=(0.3,10), (ρ,τ)=(0.6,0.5)best test accuracy (%): 73.86±2.41 -> 81.57±3.68Improved
Add label correction (ablation)FedCorr · CIFAR-10 IID, (ρ,τ)=(0.8,0.5)best test accuracy (%): 88.77±1.10 -> 90.59±0.70Improved
Use fraction scheduling (ablation)FedCorr · CIFAR-10 IID, (ρ,τ)=(0.8,0.5)best test accuracy (%): 72.71±5.03 -> 90.59±0.70Improved
Add adaptive local proximal regularization (ablation)FedCorr · CIFAR-10 IID, (ρ,τ)=(0.8,0.5)best test accuracy (%): 88.74±1.72 -> 90.59±0.70Improved
Add the finetuning stage (ablation)FedCorr · CIFAR-10 IID, (ρ,τ)=(0.8,0.5)best test accuracy (%): 83.81±2.59 -> 90.59±0.70Improved
Add the usual training stage (ablation)FedCorr · CIFAR-10 IID, (ρ,τ)=(0.8,0.5)best test accuracy (%): 88.97±1.37 -> 90.59±0.70Improved
Add mixup augmentation (ablation)FedCorr · CIFAR-10 IID, (ρ,τ)=(0.8,0.5)best test accuracy (%): 83.29±1.78 -> 90.59±0.70Improved
Use cumulative rather than per-iteration LID scores for identifying noisy clientsFedCorr · CIFAR-10 IID, label-noise client identificationseparation of clean and noisy clientsImproved
Apply the first two FedCorr stages before another FL methodFedDyn, Median, and PoC · CIFAR-10/100, multiple noisy-client ratiosbest test accuracy (%)Improved

Later papers that compared with this method

7 later accepted papers report FedCorr as a baseline in their results tables; 6 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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