AI Paper Atlas

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

Data Heterogeneity and Forgotten Labels in Split Federated Learning

Joana Tirana, Dimitra Tsigkari, David Solans Noguero, Nicolas Kourtellis

AAAI 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated semi-supervised and graph learning · 3 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 Hydra multi-head aggregated part-2b headsSFL · CIFAR-10, 80%-DL, random order, MobileNetglobal accuracy: 35±4 -> 66.5±0.1Improved
add Hydra headsSFL · CIFAR-10 MobileNet, cyclicperformance gap: decreases by up to 64.5%Improved
add Hydra headsSFL · CIFAR-10 ResNet101, cyclicglobal accuracy: increase of 112.2%Improved
aggregate heads at end of round (Hydra) instead of classic multi-head (ablation)multi-head SFL · CIFAR-10, 80%-DL, MobileNetglobal accuracy: 47±0.3 -> 66.5±0.1Improved
increase number of heads from 2 to 10 (ablation)SFL+Hydra · CIFAR-10, 80%-DL, cyclic, MobileNetglobal accuracy: 49±1 -> 65±0.1Improved
use fewer heads (2) than baseline SFL (ablation)SFL+Hydra · CIFAR-10, 80%-DL, cyclic, MobileNetglobal accuracy: 41.1±1 -> 49±1Improved
use larger heads (part-2b) instead of 2 layers (ablation)SFL+Hydra · ResNet101 CIFAR-10accuracy and PGHurt
use larger heads (part-2b) instead of 2 layers (ablation)SFL+Hydra · MobileNet CIFAR-10accuracy and PGNo clear effect
use cyclic processing order instead of random order at the serverSFL · CIFAR-10 etc.performance gap, BWImproved
reduce clients with same dominant label processed consecutively (phi 10 -> 1)SFL · CIFAR-10 MobileNet 80%-DLglobal accuracy: 41.1±1 -> 45±0.8Improved
use deeper cut layerSFL · MobileNet CIFAR-10 80%-DLaccuracy and PGImproved
add Hydra on Dirichlet and sharding partitionsSFL · CIFAR10 and TinyImageNetglobal accuracy: 50 ±0.4 -> 72 ±0.4 (CIFAR10 sharding)Improved

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