This book addresses the urgent need to shift machine learning model training and inference from centralized cloud environments to the edge, driven by the proliferation of edge devices and the associated communication, latency, security, and privacy challenges. It presents fundamental knowledge of training machine learning models, explores key challenges and issues specific to edge learning, and offers comprehensive techniques covering fundamental theory, edge learning algorithms, and system design considerations. The authors aim to equip researchers and developers with the necessary understanding to advance the field of edge intelligence, which fuses big data, edge computing, and machine learning.

This book is positioned as a foundational text for researchers and developers interested in edge learning, a critical area for enabling edge intelligence. It aims to provide a comprehensive overview of the theory, algorithms, and system design aspects necessary to overcome the limitations of traditional centralized learning models. The authors intend for the book to stimulate discussion and inspire further research by bridging the gap between big data, edge computing, and machine learning.