AI Paper Atlas

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

Is Heterogeneity Notorious? Taming Heterogeneity to Handle Test-Time Shift in Federated Learning

Yue Tan, Chen Chen, Weiming Zhuang, Xin Dong, Lingjuan Lyu, Guodong Long

NeurIPS 2023 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated learning with pretrained models · 4 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
FedICON vs FedAvg under covariate test-time shiftFedICON · Digit-5 MNISTtest accuracy: 86.18 -> 89.67Improved
FedICON vs FedAvg under covariate shiftFedICON · Digit-5 USPStest accuracy: 58.19 -> 72.13Improved
FedICON vs FedAvg under domain shiftFedICON · Office-10 Caltechtest accuracy: 26.27 -> 33.28Improved
FedICON vs FedAvg on Office-10 WebcamFedICON · Office-10 Webcamtest accuracy: 48.21 -> 48.21No clear effect
add test-time adaptation (Tent) to FedAvg-style training instead of FedICONTent · Office-10 Amazontest accuracy: 44.91 -> 38.25Hurt
personalized FL (FedRep) instead of generic modelFedRep · Digit-5 MNIST-Mtest accuracy: 40.72 -> 19.33Hurt
FedICON on CIFAR10 attribute shift vs FedAvg, Dir(0.1)FedICON · CIFAR10 Dir(0.1)attribute-shift accuracy: 50.09 -> 86.47Improved
FedICON smallest drop between original and attribute-shift testFedICON · CIFAR10 Dir(0.1)accuracy drop Delta: -14.16 -> -3.77Improved
increase number of clients from 5 to 80FedAvg and FedICON · Digit-5 covariate shiftavg test accuracyHurt
FedICON vs FedAvg across client countsFedICON · Digit-5 covariate shiftavg test accuracyImproved
increase number of local epochs from 1 to 9FedAvg · Digit-5 covariate shiftavg test accuracyHurt
increase number of local epochs from 1 to 9FedICON · Digit-5 covariate shiftavg test accuracyImproved

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