AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

When Does Differentially Private Learning Not Suffer in High Dimensions?

Xuechen Li, Daogao Liu, Tatsunori Hashimoto, Huseyin A Inan, Janardhan Kulkarni, YinTat Lee, Abhradeep Guha Thakurta

NeurIPS 2022 · Federated learning and privacy › Differential privacy · approach: Private training with public data · 8 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
Project clipped gradients onto top k=100 principal components during private fine-tuning (ablation)DP-SGD on DistilRoBERTa · SST-2 devaccuracy: 90 -> 90 (original private accuracy recovered at k=100)No clear effect
Project gradients onto fewer components (k in 10, 20) rather than 100 (ablation)DP-SGD on DistilRoBERTa · SST-2 devaccuracyMixed
Use usual classification head (random linear head on [CLS]) instead of template-completion formulation for subspace retraining (ablation)Projected DP-SGD d=100 · SST-2 devaccuracy: 90 -> 50Hurt
Increase orthogonal iterations T from 10 to 100 (ablation)Spectral analysis of gradients · DistilRoBERTa SST-2 gradientseigenspectrum decayNo clear effect
Vary number of gradient samples r in PCA (ablation)Spectral analysis of gradients · DistilRoBERTa SST-2 gradientseigenspectrum decayNo clear effect
Exclude initial gradients (use steps 300-1300 instead of first 1000) (ablation)Spectral analysis of gradients · DistilRoBERTa SST-2 gradientseigenspectrum decayNo clear effect
Scale model to Roberta-base (14M) and Roberta-large (50M trainable params)Gradient spectral analysis · SST-2 fine-tuningeigenspectrum decayNo clear effect
Increase dimension d with constant Lipschitz coefficients (A_const)DP-SGD on generalized geometric median · Synthetic geometric medianexcess empirical lossHurt
Use rapidly decaying coefficients (A_sqrt, A_linear) instead of A_constDP-SGD on generalized geometric median · Synthetic geometric medianexcess empirical loss vs dimensionImproved

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