Course

Introduction to Federated Learning – Overview of Federated Learning: Definition, History, and Applications – Concepts and Terminology – Federated Learning Architecture -Machine Learning Perspective -Security & Privacy in Federated Learning – Federated Learning vs Centralized Learning: Comparison and Contrast

Horizontal Federated Learning (HFL) -Definition and Architecture of Horizontal Federated Learning – Federated Averaging (FedAvg) Algorithm – Improvements on the FedAvg Algorithm

Vertical Federated Learning (VFL) – Definition and Architecture of Vertical Federated Learning – VFL Algorithms: Secure Federated Linear Regression, Secure Federated Tree Boosting

Federated Learning with Non-IID Data – Heterogeneity in Federated Learning -Stratification and Local Updated Rules – Advanced Optimization Techniques in Federated Learning -Adaptive Learning Rate -Momentum and Weight Decay

Federated Transfer Learning (FTL) – Framework of Federated Transfer Learning – Homomorphic Encryption in FTL – FTL Training Process -FTL Prediction Process – Security Analysis of FTL – Secret Sharing based FTL

Security in Federated Learning – Protecting Against Data Leakage in FL -Private Parameter Aggregation for FL – Data Leakage in FL Advanced Security Issues -Dealing with Byzantine Threats to Neural Networks in FL

Practical Applications and Case Studies -Real-world Applications of Federated Learning

Preamble

Federated Learning can improve the performance of models by leveraging the diversity of the data across different devices. In this course, students will learn basics of Federated Learning and will be able to apply the real-time updates of the model in various practical scenarios.

Course Objectives

• Get exposure to need for distributed model updates
• To understand the importance of privacy and security in machine learning techniques

Course Outcomes

Prerequisites

• Machine Learning

CO-PO Mapping

Evaluation Pattern – 70:30

• Midterm Exam – 20%
• Continuous Evaluation – 50%
• End Semester Exam – 30%

Text Book / References

1. Federated learning comprehensive overview of methods and applications Springer Nature Switzerland AG; 1st ed. 2022 edition By Heiko Ludwig (Editor), Nathalie Baracaldo.
2. Federated Learning, Morgan Claypool Publishers, By Qiang Yang, Yang Liu, Yong Cheng, Yan Kang, Tianjian Chen, and Han Yu.
3. Federated Learning with Python by Kiyoshi Nakayama PhD, George Jeno, O’Reilly Media, Inc. Pub.
4. What-is-federated learning? By Emily Glanz, Nova Fallen, O’Reilly Media, Inc. Pub.