This book is a teaching material for introduction to and improvement of the knowledge of machine learning and deep learning. It systematically and profoundly describes the mainstream methods and theories of machine learning and deep learning, and closely links them with engineering practice and application. This book consists of 3 parts and 21 chapters: Chapters 1~3 (Part 1) respectively introduce the basic principle of machine learning, the required mathematical knowledge (including calculus, linear algebra, probability theory and optimization methods), and the core concepts of machine learning. Chapters 4~20 (Part 2) introduce some common supervised learning algorithms, unsupervised learning algorithms, semi-supervised learning algorithms and reinforcement learning algorithms. Chapter 21 (Part 3) introduces machine learning and deep learning algorithms, and presents typical solutions to the problems cropping up in practice. Appendix A summarizes the various machine learning algorithms, Appendix B presents the evolutionary relationship of gradient descent method, and Appendix C shows the derivation of EM algorithm.

Lei Ming, graduated from Tsinghua University with a Master’s Degree in Computer Science, has over 12 years of experiences in academic research and product R&D of computer vision, machine learning and deep learning. Lei had been working at Baidu, he is now CTO and Head the Platform R&D Center of ZMODO Technology Corp., Ltd.