AI Paper Atlas

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

FedTSA: A Cluster-based Two-Stage Aggregation Method for Model-heterogeneous Federated Learning

Boyu Fan, Chenrui Wu, Xiang Su, Pan HUI

ECCV 2024 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Non-IID federated learning · 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
replace partial-training aggregation (HeteroFL) with FedTSA two-stage DML aggregationmodel-heterogeneous FL · CIFAR-100 Dir(0.3)test accuracy (%): 53.37 -> 59.57Improved
add DML knowledge exchange via diffusion data instead of HeteroFL when pruning rate is very low (ablation)FedTSA · CIFAR-10 IID, heterogeneity [0.6,0.8,1.0] to [0.1,0.4,0.7,1.0]test accuracy (%): FedTSA 89.28 -> 82.37 vs HeteroFL 82.61 -> 68.21Improved
increase DML global epochs on generated data (ablation)FedTSA · CIFAR-10 and CIFAR-100test accuracyHurt
add cross-entropy loss to KL loss in DML (ablation)FedTSA · CIFAR-10test accuracyImproved
increase synthetic images beyond 200 (ablation)FedTSA · CIFAR-10 and CIFAR-100test accuracyNo clear effect
use fewer than 150 synthetic images (ablation)FedTSA · CIFAR-10 and CIFAR-100test accuracyHurt
raise DML temperature from 5 to 20 (ablation)FedTSA · CIFAR-10 IIDtest accuracy (%): 89.28 -> 87.30Hurt
use descriptive prompt "a photo of real bird" instead of single word "bird" (ablation)diffusion data generation · CIFAR-10test accuracy (%): 81.31 -> 89.28Improved
add dataset-style detail ("in CIFAR-10 style") to the prompt (ablation)diffusion data generation · CIFAR-10test accuracy (%): 89.28 -> 89.26No clear effect
use diffusion-generated data instead of CGAN-generated data for DMLFedTSA · CIFAR-10 style imagesimage quality (qualitative)Improved
exclude slow Raspberry Pi devices from FedAvg trainingFedAvg · MNIST motivation studyconvergence time (s): 2731 -> 1428Improved

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