List your company
ResearchFormerly WizWand

Image Classification on CIFAR-100 Practical Heterogeneity (Accuracy)

59.89AccuracyFedKT-CSD
31.685239.007646.3353.6524Jul 8, 2026
Updated 4mo ago

Evaluation results

MethodLinks
FedKT-CSDHeterogeneity Type=Practical, Backbone=ResNet-18, Number of clients=20, Participation=Full, Communication rounds=1 round (synthetic pretraining) + local fine-tuning2026.0759.89
APPLEHeterogeneity Type=Practical, Backbone=ResNet-18, Number of clients=20, Participation=Full, Communication rounds=502026.0752.14
FedPHPHeterogeneity Type=Practical, Backbone=ResNet-18, Number of clients=20, Participation=Full, Communication rounds=502026.0751.21
PartialFedHeterogeneity Type=Practical, Backbone=ResNet-18, Number of clients=20, Participation=Full, Communication rounds=502026.0749.85
FedAMPHeterogeneity Type=Practical, Backbone=ResNet-18, Number of clients=20, Participation=Full, Communication rounds=502026.0749.68
pFedMeHeterogeneity Type=Practical, Backbone=ResNet-18, Number of clients=20, Participation=Full, Communication rounds=502026.0748.12
FedFomoHeterogeneity Type=Practical, Backbone=ResNet-18, Number of clients=20, Participation=Full, Communication rounds=502026.0746.21
Per-FedAvgHeterogeneity Type=Practical, Backbone=ResNet-18, Number of clients=20, Participation=Full, Communication rounds=502026.0745.38
FedAvgHeterogeneity Type=Practical, Backbone=ResNet-18, Number of clients=20, Participation=Full, Communication rounds=502026.0732.77