Scholarlink

<p>Machine Learning (ML) is rapidly advancing, enabling various applications that improve people’s work and daily lives. However, this technical progress brings privacy concerns, leading to the emergence of Privacy-Preserving Machine Learning (PPML) as a popular research topic. In this work, we investigate the privacy protection topic in ML, and showcase the advantages of Homomorphic Encryption (HE) among different privacy-preserving techniques. Additionally, this work presents an introduction of approximate HE, emphasizing its advantages and providing the detail of some representative schemes. Moreover, we systematically review the related works about approximate HE based PPML schemes from the four technical applications and three advanced applications, along with their application scenarios, models and datasets. Finally, we suggest some potential future directions to guide readers in extending the research of PPML.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Approximate homomorphic encryption based privacy-preserving machine learning: a survey

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年1月6日

查看全文
  • Jiangjun Yuan,
  • Weinan Liu,
  • Jiawen Shi,
  • Qingqing Li

摘要

Machine Learning (ML) is rapidly advancing, enabling various applications that improve people’s work and daily lives. However, this technical progress brings privacy concerns, leading to the emergence of Privacy-Preserving Machine Learning (PPML) as a popular research topic. In this work, we investigate the privacy protection topic in ML, and showcase the advantages of Homomorphic Encryption (HE) among different privacy-preserving techniques. Additionally, this work presents an introduction of approximate HE, emphasizing its advantages and providing the detail of some representative schemes. Moreover, we systematically review the related works about approximate HE based PPML schemes from the four technical applications and three advanced applications, along with their application scenarios, models and datasets. Finally, we suggest some potential future directions to guide readers in extending the research of PPML.