Foundations of Federated Learning

Erscheinungsjahr 2027
ISBN: 978-3-032-34083-2
Verlag: Springer

Buch, Englisch, 780 Seiten, Format (B × H): 155 mm x 235 mm

Reihe: Springer Series in the Data Sciences

This book provides a rigorous yet accessible introduction to federated learning as a privacy-aware, decentralized, and trustworthy approach to artificial intelligence. As data becomes increasingly distributed across devices, organizations, institutions, and jurisdictions, conventional centralized machine learning faces growing challenges related to privacy, regulation, communication costs, data ownership, and trust. The book explains the mathematical foundations, optimization methods, system architectures, and key algorithms that enable collaborative model training without centralizing sensitive data. It also examines heterogeneous data, scalability, fault tolerance, privacy and security mechanisms, governance, fairness, accountability, and regulatory compliance. Suitable for students, researchers, educators, and practitioners, it offers a unified framework for designing responsible, secure, and privacy-preserving AI systems worldwide.

Zielgruppe

Upper undergraduate

Autoren/Hrsg.

Ray, Partha Pratim

Fachgebiete

• Mathematik | Informatik EDV | Informatik Informatik Mensch-Maschine-Interaktion Informationsvisualisierung
• Mathematik | Informatik Mathematik Stochastik
• Mathematik | Informatik EDV | Informatik Informatik Mathematik für Informatiker

Inhaltsverzeichnis

Part I: Foundations of Federated Learning.- Chapter 1: Introduction to Federated Learning.- Chapter 2: Fundamentals and Definitions of Federated Learning.- Chapter 3: Mathematical Formulation of Federated Learning.- Part II: Federated Learning Architectures and Training Protocols.- Chapter 4: Federated Learning System Architectures.-Chapter 5: Decentralized and Peer-to-Peer Federated Learning.- Chapter 6: Federated Learning Training Process.- Part III: Data Heterogeneity and Non-IID Learning.- Chapter 7: Data Heterogeneity and Non-IID Challenges.- Chapter 8: Techniques for Handling Non-IID Data.- Part IV: Federated Optimization Algorithms.- Chapter 9: Foundations of Federated Optimization Algorithms.- Chapter 10: Federated Averaging.- Chapter 11: Advanced Federated Optimization
Methods.- Part V: Personalized, Heterogeneous, and Hybrid Federated Learning.- Chapter 12: Personalized Federated Learning.- Chapter 13: Heterogeneous Federated Learning.- Chapter 14: Hybrid Federated Learning.- Part VI: Privacy, Security, Governance, and Applications.- Chapter 15: Privacy-Preserving Techniques in Federated Learning.- Chapter 16: Security Threats and Robust Federated Learning.- Chapter 17: Governance, Ethics, and Incentive Mechanisms.- Chapter 18: Applications, Emerging Trends and Tools.

Über Autor(innen)

Partha Pratim Ray is an Assistant Professor (Stage 2, Academic Level 11) in the Department of Computer Applications at Sikkim University, India, with over 13 years of experience in teaching, research, and academic administration. He is currently pursuing his Ph.D. in Computer Applications, focusing on enabling large language models on resource-constrained edge systems. He holds an M.Tech in Electronics and Communication Engineering (Embedded Systems) and a B.Tech in Computer Science and Engineering, and has qualified GATE in Computer Science. His research spans Pervasive Generative AI, Internet of Things (IoT), Edge Computing, and Pervasive Biomedical Informatics, with a strong emphasis on integrating AI into real-world, resource-constrained environments. He is credited with coining and advancing the concepts of Pervasive Biomedical Informatics and the Internet of AI Things (IoAT), alongside his contributions to the emerging paradigm of Pervasive Generative AI. He has an outstanding research record with over 160 publications, including more than 120 SCI-indexed journal papers, along with books, book chapters, and numerous international conference contributions. His scholarly impact is reflected in over 14,230 citations, an h-index of 42, and an i10-index of 87 on Google Scholar, demonstrating sustained global influence and high research visibility. His work has received global recognition, reflected in his inclusion among the World’s Top 2% Scientists (Stanford University). He serves as Associate Editor for Frontiers journals, Editorial Board Member of EAI Endorsed Transactions on AI and Robotics, and actively contributes to the academic community as a reviewer, session chair, and organizer of international conferences. He is a Senior Member of IEEE and INAE, and a Fellow of IETE, reflecting his strong professional standing. His contributions also include funded projects and impactful interdisciplinary research bridging AI, IoT, edge, healthcare, and intelligent systems. He is also a visionary advocate of Indian Knowledge Systems and has informally established the Indian Knowledge Forum to promote and disseminate indigenous knowledge traditions in contemporary contexts.