En savoir plus
This book surveys normalization techniques with a deep analysis in training deep neural networks. Normalization methods can improve the training stability, optimization efficiency, and generalization ability of deep neural networks (DNNs) and have become basic components in most state-of-the-art DNN architectures. The author provides guidelines for elaborating, understanding, and applying normalization methods. This book is ideal for readers working on the development of novel deep learning algorithms and/or their applications to solve practical problems in computer vision and machine learning. The book also serves as a resource researchers, engineers, and students who are new to the field and need to understand and train DNNs. This Second Edition builds upon the original material with the addition of more recent proposed methods and expanded technical details for new normalization methods and network architectures tailored to specific tasks.
A propos de l'auteur
Lei Huang, Ph.D., is an Associate Professor and Doctoral Supervisor in the School of Artificial Intelligence (Institute) at Beihang University, China. His current research mainly focuses on normalization techniques (involving methods, theories, and applications) in training DNNs. He also has wide interests in deep learning theory (representation and optimization) and computer vision tasks. Dr. Huang has published over 40 peer-reviewed articles in top-tier machine learning and computer vision conferences and journals such as CVPR, NeurIPS, ICML, and IEEE TPAMI. He serves in program committees and as a reviewer for the top-tier conferences and journals in machine learning and computer vision. Dr. Huang received his B.Sc. and Ph.D. degrees from Beihang University.
Résumé
This book surveys normalization techniques with a deep analysis in training deep neural networks. Normalization methods can improve the training stability, optimization efficiency, and generalization ability of deep neural networks (DNNs) and have become basic components in most state-of-the-art DNN architectures. The author provides guidelines for elaborating, understanding, and applying normalization methods. This book is ideal for readers working on the development of novel deep learning algorithms and/or their applications to solve practical problems in computer vision and machine learning. The book also serves as a resource researchers, engineers, and students who are new to the field and need to understand and train DNNs. This Second Edition builds upon the original material with the addition of more recent proposed methods and expanded technical details for new normalization methods and network architectures tailored to specific tasks.
