AI Paper Atlas

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

Federated Learning Over Images: Vertical Decompositions and Pre-Trained Backbones Are Difficult to Beat

Erdong Hu, Yuxin Tang, Anastasios Kyrillidis, Chris Jermaine

ICCV 2023 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Vertical federated learning · 13 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
use frozen pre-trained backbones (ResNet101+DenseNet121 features) with an MLP instead of full backpropagation through a ResNet18 (ablation)federated image classification · Stanford Cars i.i.d.final accuracy within budget: 2.8 -> 66.0 (MOON) / 65.7 (FedProx)Improved
use FedFull (full ResNet18 training) instead of pre-trained backbone under 100 teraFLOP budget (ablation)FedProx · all six datasets, i.i.d.success reaching target accuracy: FAIL on every datasetHurt
choose among FedAvg, FedProx, MOON, FedNova, IST when only final accuracy matters on i.i.d. datafederated learning methods · i.i.d., 6 datasetsfinal accuracy: Birds 73.5-74.8No clear effect
use IST (vertical decomposition) instead of FedAvg on i.i.d. datafederated MLP head · CUB-200-2011 i.i.d.GFLOPs to threshold: 904 -> 11Improved
use IST instead of FedAvg on i.i.d. datafederated MLP head · CUB-200-2011 i.i.d.communication to threshold (GB): 9 -> 3Improved
add proximal term to FedAvg (FedProx)FedAvg · i.i.d. and skewedFLOPs to threshold: 904 -> 53 GFLOPs on Birds i.i.d.Improved
add proximal term to IST (ISTProx)IST · i.i.d. and skewedfinal accuracy and cost: 74.6 -> 74.8 on Birds i.i.d.No clear effect
use IST instead of FedAvg under extreme label skew on CIFAR-100federated learning · CIFAR-100 skewedfinal accuracy: 61.4 -> 49.4Hurt
use FedAvg instead of IST on Stanford Cars under skewfederated learning · Stanford Cars skewedfinal accuracy: 50.8 -> 56.4Improved
increase active-site participation from 2% to 6% (1000 sites) for IST/FedAdam (ablation)IST, FedAdam · skewed, 1000 sitesaccuracy change: FedAdam +12.8 on Birds; IST +1.2 on BirdsImproved
increase participation from 10% to 30% (100 sites) (ablation)FedAvg, FedProx, MOON, FedNova · skewed, 100 sitesaccuracy change: FedAvg -0.5 Birds, +4 CIFARMixed
increase number of local iterations and compare methods in FL descent direction similarity to centralized trainingIST vs FedAvg/FedProx · Flowers skewed, 20 epochscosine similarity to central descent direction: 0.01 -> 0.035Improved

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