1 Introduction
1.2 Federated Learning as a Solution
1.2.1 The Definition of Federated Learning
1.2.2 Categories of Federated Learning
1.3 Current Development in Federated Learning
1.3.1 Research Issues in Federated Learning
1.3.4 The Federated AI Ecosystem
2.1 Privacy-Preserving Machine Learning
2.3 Threat and Security Models
2.3.2 Adversary and Security Models
2.4 Privacy Preservation Techniques
2.4.1 Secure Multi-Party Computation
3 Distributed Machine Learning
3.1.2 DML Platforms
3.2 Scalability-Motivated DML
3.2.1 Large-Scale Machine Learning
3.2.2 Scalability-Oriented DML Schemes
3.3 Privacy-Motivated DML
3.3.1 Privacy-Preserving Decision Trees
3.3.2 Privacy-Preserving Techniques
3.3.3 Privacy-Preserving DML Schemes
3.4 Privacy-Preserving Gradient Descent
3.4.1 Vanilla Federated Learning
3.4.2 Privacy-Preserving Methods
3.5 Summary
4 Horizontal Federated Learning
4.1 The Definition of HFL
4.2 Architecture of HFL
4.2.1 The Client-Server Architecture
4.2.2 The Peer-to-Peer Architecture
4.2.3 Global Model Evaluation
4.3 The Federated Averaging Algorithm
4.3.1 Federated Optimization
4.3.2 The FedAvg Algorithm
4.3.3 The Secured FedAvg Algorithm
4.4 Improvement of the FedAvg Algorithm
4.4.1 Communication Efficiency
4.4.2 Client Selection
4.5 Related Works
4.6 Challenges and Outlook
5.1 The Definition of VFL
5.2 Architecture of VFL
5.3 Algorithms of VFL
5.3.1 Secure Federated Linear Regression
5.3.2 Secure Federated Tree-Boosting
5.4 Challenges and Outlook
6.1 Heterogeneous Federated Learning
6.2 Federated Transfer Learning
6.3 The FTL Framework
6.3.1 Additively Homomorphic Encryption
6.3.2 The FTL Training Process
6.3.3 The FTL Prediction Process
6.3.4 Security Analysis
