AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

Private Learning with Public Feature Conditioning

Shuli Jiang, Walid Krichene, Nicolas Mayoraz

ICML 2026 · Federated learning and privacy › Differential privacy · approach: Private training with public data · 0 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
condition public-feature embeddings with C = V Sigma^-1 V^T (Cond-DP)DPSGD · synthetic private linear regressionvalidation MSEImproved
increase spectral decay of public features (larger p) (ablation)Cond-DP vs DPSGD · synthetic private linear regressionvalidation MSEImproved
apply conditioning when singular values are uniform (p = 0) (ablation)Cond-DP · synthetic private linear regressionvalidation MSENo clear effect
condition public features with Cond-DPDPSGD · real-world linear regression (wine, Boston, energy, CA housing)validation MSEImproved
switch from conditioning to DPSGD after early epochs (Switch-Cond-DP)two-layer MLP regression · supercond, TomsHardware, wavevalidation MSEImproved
keep conditioning throughout training instead of switching to DPSGD (ablation)Cond-DP on MLP models · MLP regressionloss reductionHurt
use Cond-DP instead of DPSGDlinear model · Criteo Sponsored Search Conversionvalidation MSEImproved
use Cond-DP instead of RR-on-Binslinear model · Criteo Sponsored Search Conversionvalidation MSEMixed
use Switch-Cond-DP instead of DPSGD/RR-on-BinsMLP (128, 64) · Criteo Sponsored Search Conversionvalidation MSEMixed

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