absorb.md

Data Privacy

Dario Amodei1
No compiled wiki article for this topic yet. Raw entries below are the source material — a wiki article can be generated on demand from /admin/triggers.

Differentially Private Equivalence Testing (DP-TOST) for Sensitive Data Analysis

The DP-TOST framework enables privacy-preserving equivalence testing for means and proportions in sensitive data analysis, particularly in healthcare settings. It addresses the privacy risks of traditional equivalence testing methods by employing simulation-based calibration to maintain type-I error