AI Paper Atlas

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

Personalization Improves Privacy-Accuracy Tradeoffs in Federated Learning

Alberto Bietti, Chen-Yu Wei, Miroslav Dudik, John Langford, Steven Wu

ICML 2022 · Federated learning and privacy › Communication-efficient FL · approach: Federated optimization algorithms · 6 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 smaller α (more local personalization) instead of global learning (ablation)PP-SGD · Synthetic regressionexcess riskImproved
Use local-only learning (α=0) instead of joint learning (ablation)PP-SGD · Synthetic regression with larger privacy noiseexcess riskImproved
Use intermediate personalization (α=10) instead of local or global learning (ablation)PP-SGD · Stackoverflow tag predictiontop-1 accuracyImproved
Use smaller α instead of large α (ablation)PP-SGD · Stackoverflow tag prediction with privacy noisetop-1 accuracyImproved
Use local-only learning (α=0) instead of joint learning (ablation)PP-SGD · Federated EMNISTtest accuracyHurt
Use smaller α to increase personalization (ablation)PP-SGD · Federated EMNISTtest accuracy at good privacy guaranteesImproved
Use smaller α instead of large α (ablation)PP-SGD with a shared CNN representation · Federated EMNIST CNNtest accuracy at good privacy guaranteesImproved
Use large α instead of small α (ablation)PP-SGD with a shared CNN representation · Federated EMNIST CNNtest accuracy at poor privacy levelsImproved
Use local-only learning (α=0), leaving the shared representation untrained (ablation)PP-SGD with a shared CNN representation · Federated EMNIST CNNtest accuracyHurt
Optimize α along with privacy noise and learning rate instead of choosing only local, global, or intermediate settings (ablation)PP-SGD · Stackoverflow and EMNISTaccuracy-privacy tradeoffImproved
Sample one user per round on average instead of sampling all users each round (ablation)PPSGD with client sampling · Theoretical generalization boundbias-term convergence rate: 1/Nn instead of 1/nImproved
Optimize average user risk instead of sample-weighted risk with heterogeneous user sample sizes (ablation)PPSGD with client sampling · Heterogeneous sample sizesprivacy costImproved

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