Research map / Federated learning and privacy / Differential privacy
Differentially Private Empirical Risk Minimization under the Fairness Lens
Cuong Tran, My H Dinh, Ferdinando Fioretto
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Add gradient-disparity and decision-boundary-proximity penalties during DP-SGD | DP-SGD on Bank · Bank | excessive risk gap | Improved |
| Add gradient-disparity and decision-boundary-proximity penalties during DP-SGD | DP-SGD on Bank · Bank | excessive risk | Improved |
| Increase the input norms of a group | DP-SGD · Several datasets | excessive risk | Hurt |
| Increase the group's Hessian trace | DP-SGD noise addition · Group-level analysis | excessive risk | Hurt |
| Increase the group's Hessian trace relative to the population | Output perturbation · Group-level analysis | excessive risk gap | Hurt |
| Use equal average input norms across groups | Output perturbation with L2 loss · Group-level analysis | excessive risk gap | Improved |
| Increase the DP-SGD clipping bound | DP-SGD · Bank | gradient norm after clipping | Improved |
| Increase the clipping bound enough for noise addition to dominate clipping | DP-SGD · Group-level analysis | noise contribution to excessive risk | Improved |
Closest papers by meaning
- Optimal Differentially Private Model Training with Public Data (ICML 2024 · similarity 0.90)
- A Simple and Practical Method for Reducing the Disparate Impact of Differential Privacy (AAAI 2024 · similarity 0.89)
- Auditing Private Prediction (ICML 2024 · similarity 0.88)
- The Limits of Differential Privacy in Online Learning (NeurIPS 2024 · similarity 0.88)
- The importance of feature preprocessing for differentially private linear optimization (ICLR 2024 · similarity 0.88)
- Differential Privacy in Scalable General Kernel Learning via $K$-means Nystr{\"o}m Random Features (NeurIPS 2024 · similarity 0.88)
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.
