AI Paper Atlas

Research map / Federated learning and privacy / Communication-efficient FL

A Non-parametric View of FedAvg and FedProx:Beyond Stationary Points

Lili Su, Jiaming Xu, Pengkun Yang

JMLR 2023 · Federated learning and privacy › Communication-efficient FL · approach: Local drift correction · 20 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 more local steps s (s>=2) or FedProx instead of s=1FedAvg · Federated linear regression (M=25, d=100)gradient norm at global riskHurt
use local steps s=5,10 or FedProx instead of s=1FedAvg · Federated linear regression (M=25, d=100)estimation error ||theta_t - theta*||No clear effect
use minibatches (B=20,50,100) instead of full batchFedAvg s=5,10 · Federated linear regression (M=25, d=100)gradient magnitudeNo clear effect
use minibatches (B=20,50,100) instead of full batchFedAvg s=1 · Federated linear regression (M=25, d=100)gradient magnitudeHurt
use minibatches (B=20,50,100) instead of full batchFedAvg and FedProx · Federated linear regression (M=25, d=100)estimation errorNo clear effect
vary local steps s (1,5,10) or use FedProxFedAvg/FedProx · Federated linear regression (M=20, d=100)federation gainNo clear effect
increase model heterogeneity GammaFedAvg/FedProx · Federated linear regression (M=20, d=100)federation gain: drops below 1 at Gamma 7.5 (scarce), 0.3 (rich)Hurt
client has scarce local data (n_i=50) instead of rich (n_i=500)FedAvg/FedProx · Federated linear regression (M=20, d=100)federation gain: 7.5 vs 0.3 Gamma thresholdImproved
increase subspace dimension r of covariates to about 16-23FedAvg/FedProx · Federated linear regression (subspace covariates)federation gain: jumps from r about 16 to 23Mixed

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