Inference and Learning from Data: Volume 3: Learning

This three-volume work is a comprehensive introduction to mathematical and statistical techniques for data-driven learning and inference, with a focus on learning methods, regularization, online learning, kernel methods, feedforward and recurrent neural networks, meta-learning, and adversarial attacks. It is ideal for early-career researchers and graduate students in signal processing, machine learning, data, and inference.

Format: Hardback
Length: 990 pages
Publication date: 22 December 2022
Publisher: Cambridge University Press

This remarkable three-volume work, penned by a foremost authority in the field, presents an accessible and comprehensive introduction to the diverse range of mathematical and statistical techniques that form the foundation of contemporary methods in data-driven learning and inference. The concluding volume, Learning, builds upon the foundational topics established in volume I, offering a thorough exploration of learning methods. These methods encompass techniques such as least-squares methods, regularization, online learning, kernel methods, feedforward and recurrent neural networks, meta-learning, and adversarial attacks. A consistent structure and pedagogy are employed throughout this volume to enhance student understanding, with over 350 end-of-chapter problems (including complete solutions for instructors), 280 figures, 100 solved examples, datasets, and downloadable Matlab code. Supported by sister volumes Foundations and Inference, and unparalleled in its scope and depth, this textbook sequence is an ideal resource for early-career researchers and graduate students pursuing courses in signal processing, machine learning, data, and inference.

ISBN-13: 9781009218283