PulseAugur
LIVE 15:37:58
research · [2 sources] ·
0
research

New CorrDP framework relaxes differential privacy for correlated features

Researchers have developed a new framework called CorrDP that modifies differential privacy to account for feature correlations. This approach allows for relaxed privacy constraints on insensitive features, even if they are correlated with sensitive ones, by quantifying these correlations using total variation distance. The framework includes algorithms for differentially private empirical risk minimization (DP-ERM) that use distance-dependent noise in gradients, offering improved theoretical utility guarantees. Experiments on synthetic and real-world data demonstrate that CorrDP-based DP-ERM outperforms standard DP methods when insensitive features are present. AI

Summary written by gemini-2.5-flash-lite from 2 sources. How we write summaries →

IMPACT Introduces a more nuanced approach to differential privacy, potentially improving utility in machine learning models with correlated features.

RANK_REASON Academic paper introducing a new framework for differential privacy.

Read on arXiv cs.LG →

COVERAGE [2]

  1. arXiv cs.LG TIER_1 · Tianyu Wang, Luhao Zhang, Rachel Cummings ·

    Integrating Feature Correlation in Differential Privacy with Applications in DP-ERM

    arXiv:2605.03945v1 Announce Type: new Abstract: Standard differential privacy imposes uniform privacy constraints across all features, overlooking the inherent distinction between sensitive and insensitive features in practice. In this paper, we introduce a relaxed definition of …

  2. arXiv cs.LG TIER_1 · Rachel Cummings ·

    Integrating Feature Correlation in Differential Privacy with Applications in DP-ERM

    Standard differential privacy imposes uniform privacy constraints across all features, overlooking the inherent distinction between sensitive and insensitive features in practice. In this paper, we introduce a relaxed definition of differential privacy that accounts for such hete…