Neural Networks and Statistical Learning

3rd Third 2027 Auflage
ISBN: 978-1-4471-7633-6
Verlag: Springer

Buch, Englisch, Format (B × H): 155 mm x 235 mm

This book, now a fully updated and expanded third edition provides a comprehensive and accessible introduction to neural networks and statistical learning, two of the most important foundations of modern artificial intelligence. It presents a unified view of machine learning methods, combining classical statistical learning techniques with modern deep learning architectures and emerging AI technologies. Now in its third, fully revised and updated edition.

This book begins with the mathematical and theoretical foundations of machine learning, including optimization, learning theory, and statistical principles. It then introduces core neural network models and learning algorithms, covering topics such as multilayer perceptrons, recurrent neural networks, clustering methods, kernel learning, support vector machines, and probabilistic models. Building on these fundamentals, this book explores modern developments in artificial intelligence, including deep learning, graph neural networks, transformers, large language models, federated learning, and generative AI.

A distinctive feature of the book is its integrated treatment of theory, algorithms, and modern applications. Numerous illustrations, tables, and structured explanations are used to clarify complex concepts and provide readers with a clear understanding of how different learning methods are related. This new edition contains new advanced in artificial intelligence, including deep neural networks, graph neural networks, transformers, large language model AI (LLMs), federated learning, and generative AI (GenAI).

By presenting both classical approaches and recent advances within a single coherent framework, this book offers readers a systematic understanding of machine learning techniques and their role in modern intelligent systems. It serves as a valuable resource for students, researchers, and professionals working in artificial intelligence, data science, and machine learning.

Zielgruppe

Graduate

Autoren/Hrsg.

Du, Ke-Lin

Swamy, M. N. S.

Fachgebiete

• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz Neuronale Netzwerke
• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz Maschinelles Lernen

Inhaltsverzeichnis

Introduction.- Reasoning, learning paradigms, and machine teaching.- Mathematical foundations of machine learning.- Learning theory, generalization, and neural computation.- Elements of computational learning theory.- Perceptrons, multilayer perceptrons, and backpropagation learning.- Radial basis function networks.- Hopfield networks and energy-based search methods.- Boltzmann machines and energy-based learning.- Neural associative memories.- Recurrent neural networks.- Clustering models and algorithms.- Principal component analysis and subspace learning.- Nonnegative and sketch-based low-rank matrix factorization.- Independent component analysis and blind source separation.- Discriminant analysis.- Sparse signal recovery and learning.- Low-rank matrix recovery and completion.- Tensor methods for multilinear representation, recovery, and learning.- Kernel learning methods.- Support vector machines.- Bayesian networks and graphical inference.- Combining learners: Ensemble learning and data fusion.- Reinforcement learning.- Transfer learning and knowledge transfer.- Federated learning.- Spike-based neural learning.- Physics-informed neural networks and neural operators.- Deep learning: Foundations, architectures, and robustness.- Transformers: Attention, efficiency, and applications.- Generative AI: LLMs, diffusion, agents, and reliability.- Graph neural networks: Architectures, reasoning, and trustworthiness.- Fuzzy sets, approximate reasoning, and inference systems.- Neurofuzzy learning systems.- Neural hardware and parallel learning.- Data, web, and multimedia mining.

Über Autor(innen)

Ke-Lin Du is Chief Scientist at Guangzhou Digital Technology Group, China. Previously, he was a Professor at Guangdong University of Science and Technology (2024–2025) and an Affiliate Professor at Concordia University, Canada (2011–2024). He has also held positions at Huawei Technologies, the China Academy of Telecommunication Technology, the Chinese University of Hong Kong, the Hong Kong University of Science and Technology, Concordia University, Enjoyor Inc., and Xonlink Inc. He has coauthored six books, published 60 papers, and holds over 40 patents. An IEEE Senior Member and IEEE CTSoc MDA Technical Commitee Member, his research interests include signal processing, wireless communications, and machine learning. He has been listed among the World's Top 2% Most-Cited Scientists for nine consecutive years (2017–2025).

M. N. S. Swamy is a Research Professor and Concordia Chair (Tier I) in Signal Processing at Concordia University, Canada, where he previously served as Founding Chair of Electrical Engineering and Dean of Engineering and Computer Science. He has coauthored nine books, holds five patents, and supervised over 100 graduate theses. He has served as Editor-in-Chief of Circuits, Systems and Signal Processing since 1999 and as President of the IEEE Circuits and Systems Society. An IEEE Life Fellow and Fellow of several other professional organizations, he has received numerous honors, including the IEEE Mac Van Valkenburg Award (2022) and IEEE Canada McNaughton Award (2025). His research interests include circuits, systems, and signal processing.