An Introduction to Machine Learning

This textbook offers a comprehensive introduction to Machine Learning 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.

Table of Contents

1 Ambitions and Goals of Machine Learning

2 Probabilities: Bayesian Classifiers

3 Similarities: Nearest-Neighbor Classifiers

4 Inter-Class Boundaries: Linear and Polynomial Classifiers

5 Decision Trees

6 Artificial Neural Networks

7 Computational Learning Theory

8 Experience from Historical Applications

9 Voting Assemblies and Boosting

10 Classifiers in the Form of Rule-Sets

11 Practical Issues to Know About

12 Performance Evaluation

13 Statistical Significance

14 Induction in Multi-label Domains

15 Unsupervised Learning

16 Deep Learning

17 Reinforcement Learning: N-Armed Bandits and Episodes

18 Reinforcement Learning: From TD(O) to Deep-Q-Learning

19 Temporal Learning

20 Hidden Markov Models

21 Genetic Algorithm

About the Author

Miroslav Kubat, Associate Professor at the University of Miami, has been teaching and studying machine learning for over 25 years. He has published more than 100 peer-reviewed papers, co-edited two books, served on the program committees of over 60 conferences and workshops, and is an editorial board member of three scientific journals. He is widely credited with co-pioneering research in two major branches of the discipline: induction of time-varying concepts and learning from imbalanced training sets. He also contributed to research in induction from multi-label examples, induction of hierarchically organized classes, genetic algorithms, and initialization of neural networks. Professor Kubat is also known for his many practical applications of machine learning, ranging from oil-spill detection in radar images to text categorization to tumor segmentation in MR images.