Deep Learning: Foundations and Concepts aims to offer both newcomers to machine learning and those already experienced in the field a comprehensive grasp of fundamental ideas underpinning deep learning. Covering key concepts related to contemporary deep learning architectures and techniques, this essential book will equip readers with a robust foundation for potential future specialization. The field of deep learning is undergoing rapid evolution. Rather than summarizing the latest research developments, Bishop distills the key ideas in order to ensure that the foundations and concepts presented in this book will endure the test of time. For enhanced accessibility, the book is organized into numerous bite-sized chapters, each exploring a distinct topic. The narrative follows a linear progression, with each chapter building upon content from its predecessors. This structure lends itself effectively to teaching a two-semester undergraduate or postgraduate machine learning course, while remaining equally relevant to those engaged in active research or in self-study. To fully grasp machine learning, a certain level of mathematical understanding is required. The book provides a self-contained introduction to probability theory, and includes appendices summarizing useful results in linear algebra, calculus of variations, and Lagrange multipliers. However, the focus of the book is on conveying a clear understanding of ideas rather than mathematical rigor, with emphasis on real-world practical value of techniques rather than abstract theory. Complex concepts are presented from multiple perspectives including textual descriptions, diagrams, mathematical formulae, and pseudo-code to cater to readers from diverse backgrounds. This book can be viewed as a successor to Neural Networks for Pattern Recognition (Bishop, 1995a) which provided the first comprehensive treatment of neural networks from a statistical perspective. It can be considered as a companion volume to Pattern Recognition and Machine Learning (Bishop, 2006) which covered a broader range of topics in machine learning but predates the deep learning revolution.