REAEDP: Entropy-Calibrated Differentially Private Data Release with Formal Guarantees and Attack-Based Evaluation Corresponding author:Bo Ma
Bo Ma 1 , 2 Jinsong Wu 3 , 4 Wei Qi Yan 1 · 2026
Read REAEDP: Entropy-Calibrated Differentially Private Data Release with Formal Guarantees and Attack-Based Evaluation Corresponding author:Bo Ma by Bo Ma 1 , 2 Jinsong Wu 3 , 4 Wei Qi Yan 1 (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-13709.txt.
Description
Bo Ma, Jinsong Wu, and Wei Qi Yan propose REAEDP, a differentially private data release framework with formal guarantees. Data release faces output-side attacks, such as membership inference, attribute inference, and record linkage. REAEDP joins entropy-calibrated histogram release, a synthetic-data mechanism, and attack-based evaluation. They bound how much Shannon entropy can change between adjacent histogram datasets, with an extension to Rényi entropy. That bound lets noise be calibrated for private release. On public datasets, attacks fell toward random-guess levels.
Catalog record
| CREATOR | Bo Ma 1 , 2 Jinsong Wu 3 , 4 Wei Qi Yan 1 |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 68458 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.13709