This article reviews privacy challenges in machine learning and provides a critical overview of the relevant research literature. The possible adversarial models are discussed, a wide range of attacks related to sensitive information leakage is covered, and several open problems are highlighted.

中文翻译：

---

机器学习中隐私的重要概述

本文回顾了机器学习中的隐私挑战，并对相关研究文献进行了重要概述。讨论了可能的对抗模型，涵盖了与敏感信息泄漏相关的广泛攻击，并强调了几个未解决的问题。