AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

Improving the Privacy and Practicality of Objective Perturbation for Differentially Private Linear Learners

Rachel Emily Redberg, Antti Koskela, Yu-Xiang Wang

NeurIPS 2023 · Federated learning and privacy › Differential privacy · approach: DP-SGD refinements

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 Algorithm 1 instead of honest DP-SGDlogistic regression, eps=0.1 · Synthetic-Ltest accuracy (%): 77.34% -> 90.05%Improved
Use Algorithm 1 instead of honest DP-SGDlogistic regression, eps=1 · Synthetic-Ltest accuracy (%): 93.15% -> 94.50%Improved
Use Algorithm 1 instead of honest DP-SGDlogistic regression, eps=8 · Synthetic-Ltest accuracy (%): 95.17 % -> 95.30%Improved
Use Algorithm 1 instead of honest DP-SGDlogistic regression, eps=0.1 · Adulttest accuracy (%): 78.32% -> 81.37%Improved
Use Algorithm 1 instead of honest DP-SGDlogistic regression, eps=1 · Adulttest accuracy (%): 82.40% -> 83.18%Improved
Use Algorithm 1 instead of honest DP-SGDlogistic regression, eps=8 · Adulttest accuracy (%): 83.66% -> 83.99%Improved
Account for privacy cost of hyperparameter tuning (honest vs dishonest DP-SGD)DP-SGD · Synthetic-L, eps=0.1test accuracy (%): 93.85% -> 77.34%Hurt
Account for privacy cost of hyperparameter tuning (honest vs dishonest DP-SGD)DP-SGD · Adult, eps=0.1test accuracy (%): 81.61% -> 78.32%Hurt
Account for privacy cost of hyperparameter tuning (honest vs dishonest DP-SGD)DP-SGD · Synthetic-L, eps=8test accuracy (%): 95.43% -> 95.17 %Hurt
Dishonest DP-SGD (no tuning cost) compared with Algorithm 1DP-SGD vs Algorithm 1 · Synthetic-L, eps=0.1test accuracy (%): 93.85% -> 90.05%Hurt
Dishonest DP-SGD (no tuning cost) compared with Algorithm 1DP-SGD vs Algorithm 1 · Adult, eps=1test accuracy (%): 83.54% -> 83.18%Hurt
Increase privacy budget epsilon from 0.1 to 8 (gap between Algorithm 1 and honest DP-SGD narrows)Algorithm 1 vs honest DP-SGD · Synthetic-Ltest accuracy (%): 90.05% -> 95.30% vs 77.34% -> 95.17 %Mixed

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