BriefGPT.xyz
Sep, 2021
当差分隐私遇上自然语言处理：魔鬼藏在细节中
When differential privacy meets NLP: The devil is in the detail
HTML
PDF
Ivan Habernal
TL;DR
本文针对ADePT的形式化分析表明其不具备差分隐私性质，证明了实验结果缺乏可靠的支持，同时阐明了形式化保证在自然语言处理中的重要性。
Abstract
differential privacy
provides a formal approach to privacy of individuals. Applications of
differential privacy
in various scenarios, such as protecting users' original utterances, must satisfy certain mathematic
→