Federated Edge Learning
Description
This book presents various effective schemes from the perspectives of algorithms, architectures, privacy, and security to enable scalable and trustworthy Federated Edge Learning (FEEL). From the algorithmic perspective, the authors elaborate various federated optimization algorithms, including zeroth-order, first-order, and second-order methods. There is a specific emphasis on presenting provable convergence analysis to illustrate the impact of learning and wireless communication parameters. The convergence rate, computation complexity and communication overhead of the federated zeroth/first/second-order algorithms over wireless networks are elaborated.
From the networking architecture perspective, the authors illustrate how the critical challenges of FEEL can be addressed by exploiting different architectures and designing effective communication schemes. Specifically, the communication straggler issue of FEEL can be mitigated by utilizing reconfigurable intelligent surface and unmanned aerial vehicle to reconfigure the propagation environment, while over-the-air computation is utilized to support ultra-fast model aggregation for FEEL by exploiting the waveform superposition property. Additionally, the multi-cell architecture presents a feasible solution for collaborative FEEL training among multiple cells. Finally, the authors discuss the challenges of FEEL from the privacy and security perspective, followed by presenting effective communication schemes that can achieve differentially private model aggregation and Byzantine-resilient model aggregation to achieve trustworthy FEEL.
This book is designed for researchers and professionals whose focus is wireless communications. Advanced-level students majoring in computer science and electrical engineering will also find this book useful as a reference.
More details
Other editions
Additional editions
Persons
Yong Zhou received the BSc and MEng degrees from Shandong University, Jinan, China, in 2008 and 2011, respectively, and the PhD degree from the University of Waterloo, Waterloo, ON, Canada, in 2015. From Nov 2015 to Jan 2018, he worked as a postdoctoral research fellow in the Department of Electrical and Computer Engineering, The University of British Columbia, Vancouver, Canada. Since Mar. 2018, he has been with the School of Information Science and Technology, ShanghaiTech University, Shanghai, China, where he is currently a Tenured Associate Professor. His research interests include 6G communications, edge intelligence, and Internet of Things.
Wenzhi Fang received his B.S. degree from Shanghai University in 2020 and completed his master's degree at ShanghaiTech University in 2023. His research interests focus on optimization theory and its application in machine learning.
Yuanming Shi received the B.S. degree in electronic engineering from Tsinghua University, Beijing, China, in 2011. He received the Ph.D. degree in electronic and computer engineering from The Hong Kong University of Science and Technology (HKUST), in 2015. Since September 2015, he has been with the School of Information Science and Technology in ShanghaiTech University, where he is a Full Professor. His research areas include edge artificial intelligence and large-scale optimization. He is a recipient of the IEEE Marconi Prize Paper Award in Wireless Communications in 2016, the Young Author Best Paper Award by the IEEE Signal Processing Society in 2016, the IEEE ComSoc Asia-Pacific Outstanding Young Researcher Award in 2021, the Chinese Institute of Electronics First Prize in Natural Science in 2022, and the China Institute of Communications First Prize in Natural Science in 2024. He is an IET Fellow.
Khaled B. Letaief is a globally recognized leader in wireless communications and networks, with a research focus that spans artificial intelligence, integrated sensing and communication, mobile cloud and edge computing, federated learning, and 6G systems. Dr. Letaief is a distinguished member of several esteemed organizations, including the United States National Academy of Engineering, IEEE Fellow, and Fellow of the Hong Kong Institution of Engineers. He is also a member of the Hong Kong Academy of Engineering. His accolades include numerous prestigious awards, such as the 2024 IEEE James Evans Avant Garde Award, 2024 Distinguished Purdue University Alumni Award, 2022 IEEE Edwin Howard Armstrong Achievement Award, and 2021 IEEE Communications Society Best Survey Paper Award. He has also received the 2019 Joint Paper Award from the IEEE Communications Society and Information Theory Society, the 2016 IEEE Marconi Prize Award in Wireless Communications, and over 20 IEEE Best Paper Awards.
Since 1993, Dr. Letaief has been a faculty member at The Hong Kong University of Science and Technology (HKUST), where he has held multiple leadership roles, including Senior Advisor to the President, Acting Provost, Head of the Electronic and Computer Engineering Department, Director of the Wireless IC Design Center, and Director of the Hong Kong Telecom Institute of Information Technology. He served as Chair Professor and Dean of Engineering at HKUST and, from 2015 to 2018, was Provost at Hamad Bin Khalifa University in Qatar, where he played a key role in establishing a research-intensive university in collaboration with renowned institutions like Northwestern University, Carnegie Mellon University, Cornell, and Texas A&M. He earned his B.S. degree with distinction in Electrical Engineering from Purdue University in December 1984, followed by an M.S. and Ph.D. in Electrical Engineering from the same institution in August 1986 and May 1990, respectively. In 2022, he received an honorary Ph.D. from the University of Johannesburg, South Africa.
Content
Part 1: Introduction and Overview.- 1. Introduction and overview.- 1.1. Overview of federated edge learning (FEEL).- 1.2. Learning models and algorithms of FEEL.- 1.3. Motivation and challenges of FEEL.- 1.4. Organization.- Part 2: Algorithms.- 2. First-order optimization for FEEL.- 2.1. Background and motivation.- 2.2. Federated first-order optimization model and algorithm.- 2.3. Sparse and low-rank optimization for FEEL.- 2.4. Simulations and discussions.- 2.5. Summary.- 3. Second-order optimization for FEEL.- 3.1. Background and motivation.- 3.2. Federated second-order optimization model and algorithm.- 3.3. Convergence analysis.- 3.4. System optimization.- 3.5. Simulations and discussions.- 3.6. Summary.- 4. Zeroth-order optimization for FEEL.- 4.1. Background and motivation.- 4.2. Federated zeroth-order optimization model and algorithm.- 4.3. Convergence analysis.- 4.4. Over-the-air federated zeroth-order optimization.- 4.5. Simulations and discussions.- 4.6. Summary.- Part 3: Architectures.- 5. Reconfigurable intelligent surface assisted FEEL.- 5.1. Background and motivation.- 5.2. Communication and learning models.- 5.3. Convergence analysis and problem formulation.- 5.4. Alternating optimization algorithm design.- 5.5. GNN-based learning algorithm design.- 5.6. Simulations and discussions.- 5.7. Summary.- 6. Unmanned aerial vehicle assisted FEEL.- 6.1. Background and motivation.- 6.2. Communication and learning models.- 6.3. Convergence analysis and problem formulation.- 6.4. Joint device scheduling, time allocation, and trajectory design.- 6.5. Simulations and discussions.- 6.6. Summary.- 7. FEEL over multi-cellwireless networks.- 7.1. Background and motivation.- 7.2. Communication and learning models.- 7.3. Convergence analysis and problem formulation.- 7.4. Cooperative optimization for multi-cell FEEL.- 7.5. Simulations and discussions.- 7.6. Summary.- Part 4: Trustworthiness.- 8. Differentially-private FEEL.- 8.1. Background and motivation.- 8.2. System model.- 8.3. Performance analysis and privacy preserving mechanism.- 8.4. Two-step alternating low-rank optimization.- 8.5. Simulations and discussions.- 8.6. Summary.- 9. Trustworthy FEEL via blockchain.- 9.1. Background and motivation.- 9.2. System model.- 9.3. Latency analysis and problem formulation.- 9.4. TD3 based resource allocation.- 9.5. Simulations and discussions.- 9.6. Summary.- Part 5: Conclusions and Future Directions.- 10. Conclusions and future directions.- 10.1. Conclusions.- 10.2. Future directions.
