Scholarlink

This chapter introduces federated learning (FL), a distributed machine learning approach designed to address data privacy, security, and processing challenges in the era of big data and AI. FL allows model training across multiple devices or organizations without sharing raw data, thus enhancing privacy by only exchanging model updates. The chapter covers three FL modes: Horizontal Federated Learning (HFL), Vertical Federated Learning (VFL), and Transfer Federated Learning (TFL). It also explores recommender systems (RecSys), focusing on traditional methods like collaborative filtering and advanced techniques, such as graph-based approaches. Finally, the chapter discusses Federated Recommender Systems (FedRec), which integrate FL and recommender systems to improve privacy-preserving recommendation services while addressing challenges like data heterogeneity and communication overhead.

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

Introduction

  • 章节

  • 发表日期：

    2025年

查看全文
  • Xiangjie Kong,
  • Lingyun Wang,
  • Mengmeng Wang,
  • Guojiang Shen

摘要

This chapter introduces federated learning (FL), a distributed machine learning approach designed to address data privacy, security, and processing challenges in the era of big data and AI. FL allows model training across multiple devices or organizations without sharing raw data, thus enhancing privacy by only exchanging model updates. The chapter covers three FL modes: Horizontal Federated Learning (HFL), Vertical Federated Learning (VFL), and Transfer Federated Learning (TFL). It also explores recommender systems (RecSys), focusing on traditional methods like collaborative filtering and advanced techniques, such as graph-based approaches. Finally, the chapter discusses Federated Recommender Systems (FedRec), which integrate FL and recommender systems to improve privacy-preserving recommendation services while addressing challenges like data heterogeneity and communication overhead.