Abstract
Noise for differential privacy is often generated to obfuscate raw statistical results to hide identifies of specific individuals in data. Traditional methods are often inaccurate and/or computationally inefficient. In this work, a software library or program uses a differential privacy noise generator to generate high quality random numbers from the geometric distribution to both accurately and efficiently generate noise for differential privacy.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Krueger, Brett, "A Method to Efficiently Generate Noise for Differential Privacy", Technical Disclosure Commons, (April 07, 2020)
https://www.tdcommons.org/dpubs_series/3102
COinS