AI Paper Atlas

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

Improving Generalization in Federated Learning by Seeking Flat Minima

Debora Caldarola, Barbara Caputo, Marco Ciccone

ECCV 2022 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Non-IID federated learning · 110 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 ASAM local trainingFedAvg · CIFAR100 (alpha=0, 5 clients/round)accuracy (%): 30.25 -> 36.04Improved
add SAM local trainingFedAvg · CIFAR100 (alpha=0, 5 clients/round)accuracy (%): 30.25 -> 31.04Improved
add ASAM local trainingFedAvg · CIFAR10 (alpha=0, 5 clients/round)accuracy (%): 65.00 -> 73.66Improved
add server-side SWA on top of ASAMFedASAM · CIFAR100 (alpha=0, 5 clients/round)accuracy (%): 36.04 -> 42.01Improved
add server-side SWA on top of ASAMFedASAM · CIFAR10 (alpha=0, 5 clients/round)accuracy (%): 73.66 -> 76.44Improved
add SWA to FedAvgFedAvg · CIFAR100 (alpha=0, 5 clients/round)accuracy (%): 30.25 -> 39.34Improved
add ASAM in homogeneous settingFedAvg · CIFAR100 (alpha=1000, 5 clients/round)accuracy (%): 49.92 -> 54.81Improved
use 2 local epochs instead of 1FedAvg · CIFAR100 (alpha=0, 5 clients/round)accuracy (%): 30.25 -> 24.94Hurt
add ASAM local training (flatness)FedAvg · CIFAR100 Hessian (alpha=0)Hessian lambda_max: 93.46 -> 30.11Improved
add SWA on top of FedASAM (flatness)FedASAM · CIFAR100 Hessian (alpha=0)Hessian lambda_max: 30.11 -> 24.57Improved
add Mixup augmentationFedAvg · CIFAR100 (alpha=0, 5 clients/round)accuracy (%): 30.25 -> 29.91Hurt
add Cutout augmentationFedAvg · CIFAR100 (alpha=0, 5 clients/round)accuracy (%): 30.25 -> 24.24Hurt

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