A one-to-one correspondence maps maximal LDP channels under the Blackwell order to vertices of a finite-dimensional polytope, making optimal privacy-utility trade-offs computable via linear programming or vertex enumeration for general problems.
Title resolution pending
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
years
2026 6verdicts
UNVERDICTED 6representative citing papers
Trade-off functions between two distributions are finitely testable if and only if their Neyman-Pearson rejection regions are attainable by a VC-class of sets.
QOP achieves (ε, δ)-differential privacy for ERM in the interpolation regime under weaker assumptions than linear objective perturbation by using random quadratic curvature to enforce stability and control sensitivity.
Privacy and fairness cannot both be guaranteed in facility location over all datasets, but mechanisms exist that are optimal or near-optimal on welfare and fairness for natural data while preserving worst-case differential privacy.
ICSA uses invariant coordinate selection for robust latent space anonymization, outperforming spectral anonymization under outliers in simulations and clinical data while maintaining utility.
Hybrid DP with LLM or NER preprocessing significantly improves the privacy-utility trade-off for Dutch clinical note de-identification compared to standalone DP.
citing papers explorer
-
Optimal Privacy-Utility Trade-Offs in LDP: Functional and Geometric Perspectives
A one-to-one correspondence maps maximal LDP channels under the Blackwell order to vertices of a finite-dimensional polytope, making optimal privacy-utility trade-offs computable via linear programming or vertex enumeration for general problems.
-
When Are Trade-Off Functions Testable from Finite Samples?
Trade-off functions between two distributions are finitely testable if and only if their Neyman-Pearson rejection regions are attainable by a VC-class of sets.
-
Quadratic Objective Perturbation: Curvature-Based Differential Privacy
QOP achieves (ε, δ)-differential privacy for ERM in the interpolation regime under weaker assumptions than linear objective perturbation by using random quadratic curvature to enforce stability and control sensitivity.
-
Tradeoffs in Privacy, Welfare, and Fairness for Facility Location
Privacy and fairness cannot both be guaranteed in facility location over all datasets, but mechanisms exist that are optimal or near-optimal on welfare and fairness for natural data while preserving worst-case differential privacy.
-
Data anonymization in the presence of outliers via invariant coordinate selection
ICSA uses invariant coordinate selection for robust latent space anonymization, outperforming spectral anonymization under outliers in simulations and clinical data while maintaining utility.
-
Differentially Private De-identification of Dutch Clinical Notes: A Comparative Evaluation
Hybrid DP with LLM or NER preprocessing significantly improves the privacy-utility trade-off for Dutch clinical note de-identification compared to standalone DP.