AI Paper Atlas

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

Handling Data Heterogeneity via Architectural Design for Federated Visual Recognition

Sara Pieri, Jose Renato Restom, Samuel Horváth, Hisham Cholakkal

NeurIPS 2023 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Vertical federated learning · 2 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
Replace ResNet-50 with CAFormer-S18FedAvg on CIFAR-10 Split-3 · CIFAR-10 Split-3accuracy: 59.9 \pm 1.87 -> 97.1 \pm 0.14Improved
Replace ResNet-50 with ViT-Sfederated visual recognition · GLD-23Kaccuracy: 54.3 \pm 0.48 -> 76.6 \pm 0.69Improved
Replace ResNet-50 with CAFormer-S18federated visual recognition · Fed-ISIC2019accuracy: 67.6 \pm 1.59 -> 83.1 \pm 0.49Improved
Replace BatchNorm with LayerNorm (ablation)CoAtNet-0 · CIFAR-10 Split-3accuracy: 20.9 \pm 0.45 -> 94.2 \pm 0.06Improved
Replace BatchNorm with LayerNorm (ablation)PoolFormer-S12 · CIFAR-10 Split-3accuracy: 51.5 \pm 0.32 -> 91.5 \pm 0.13Improved
Replace BatchNorm with LayerNorm (ablation)ResNet-50 · CIFAR-10 Split-3accuracy: 59.9\pm 1.87 -> 72.2 \pm 3.06Improved
Use FedBN while retaining BatchNorm (ablation)ResNet-50 · CIFAR-10 Split-3accuracy: 59.9\pm 1.87 -> 59.6 \pm 1.40No clear effect
Replace BatchNorm with LayerNorm (ablation)PoolFormer-S12 · Fed-ISIC2019accuracy: 64.2 \pm 0.53 -> 83.2 \pm 0.63Improved
Use GroupNorm instead of LayerNorm (ablation)PoolFormer-S12 · CIFAR-10 Split-3accuracy: 91.5 \pm 0.13 -> 81.5 \pm 0.29Hurt
Use GroupNorm instead of LayerNorm (ablation)ResNet-50 · CIFAR-10 Split-3accuracy: 72.2 \pm 3.06 -> 89.3\pm 0.59Improved
Use FedAVGM instead of FedAVG (ablation)ResNet-50 · CIFAR-10 Split-3accuracy: 59.9 \pm 1.87 -> 80.2 \pm 0.49Improved
Use SCAFFOLD instead of FedAVG (ablation)CAFormer-S18 · Fed-ISIC2019accuracy: 83.1 \pm 0.49 -> 77.1 \pm 0.61Hurt

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