Название: An Introduction to Machine Learning, Third Edition
Автор: Miroslav Kubat
Издательство: Springer
Год: 2021
Страниц: 458
Язык: английский
Формат: pdf (true), epub
Размер: 18.3 MB

Этот учебник предлагает всестороннее введение в методы и алгоритмы машинного обучения (ML). Это третье издание охватывает новые подходы, которые стали очень актуальными, включая глубокое обучение и автоматическое кодирование, вводную информацию о темпоральном обучении и скрытых марковских моделях, а также гораздо более подробное описание обучения с подкреплением. Книга написана в простой для понимания манере, с множеством примеров и изображений, а также с множеством практических советов и обсуждений простых приложений.

This textbook offers a comprehensive introduction to Machine Learning (ML) techniques and algorithms. This Third Edition covers newer approaches that have become highly topical, including deep learning, and auto-encoding, introductory information about temporal learning and hidden Markov models, and a much more detailed treatment of reinforcement learning. The book is written in an easy-to-understand manner with many examples and pictures, and with a lot of practical advice and discussions of simple applications.

The main topics include Bayesian classifiers, nearest-neighbor classifiers, linear and polynomial classifiers, decision trees, rule-induction programs, artificial neural networks, support vector machines, boosting algorithms, unsupervised learning (including Kohonen networks and auto-encoding), deep learning, reinforcement learning, temporal learning (including long short-term memory), hidden Markov models, and the genetic algorithm. Special attention is devoted to performance evaluation, statistical assessment, and to many practical issues ranging from feature selection and feature construction to bias, context, multi-label domains, and the problem of imbalanced classes.

No less important is reinforcement learning. Whereas the book’s previous editions only mentioned the related techniques in passing, it has now become necessary to devote to them no less than two full-length chapters. Some recent achievements are so spectacular that the textbook would seem ridiculously incomplete without adequate treatment of the relevant algorithms.

Unsupervised learning, too, has gained in importance, especially those mechanisms that from existing attributes create higher-level features to describe training examples. Also mechanisms capable of visualizing multidimensional data have become quite important. An entire new section dealing with auto-encoding had to be added. Most of the other sections of this chapter were expanded. One new chapter now focuses on temporal learning. Even a beginner needs to know the basic principles of recurrent neural networks and needs to have an idea of what is long short-term memory. Another chapter introduces the reader into the realm of Hidden Markov Models. True, this is somewhat advanced, but the reader must at least know that it exists.

The older topics all remain, but in this new edition, the author sought to improve the clarity of exposition: to make the text easier to read and the pictures more pleasant to look at. These re-written chapters include the Bayesian rule, the nearest-neighbor principle, linear and polynomial classifiers, decision trees, artificial neural networks, and the boosting algorithms. Significant space is devoted to practical aspects of concrete engineering applications and to the ways of assessing their performance, including statistical evaluation.