Machine Learning: A Probabilistic Perspective by Kevin Murphy

Book Contents :-
PART I: MACHINE LEARNING: THE BASICS 1. Introduction 2. Nearest Neighbour Classification 3. Linear Dimension Reduction 4. Generalisation 5. Linear Discriminant Analysis 6. Linear Parameter Models 7. Layered Neural Networks 8. Autoencoders 9. Data Visualisation PART II: INFERENCE AND LEARNING IN PROBABILISTIC MODELS 10. Basic Concepts in Probability 11. Introducing Graphical Models 12. Inference in Belief Networks 13. The Junction Tree Algorithm 14. Variational Learning and EM PART III: PROBABILISTIC MODELS IN MACHINE LEARNING 15. Introduction to Bayesian Methods 16. Bayesian Regression 17. Logistic Regression 18. Naive Bayes 19. Mixture Models: Discrete Hidden Variables 20. Factor Analysis and PPCA 21. Independent Component Analysis 22. Dynamic Bayesian Networks: Discrete Hidden Variables 23. Dynamic Continuous Hiddens: Linear Dynamical Systems 24. Switching Linear Dynamical Systems 25. Gaussian Processes PART IV: APPROXIMATE INFERENCE METHODS 26. Sampling APPENDICES A. Mathematics B. Graph Terminology C. Some Standard Distributions D. Bounds on Convex Functions E. Positive Definite Matrices and Kernel Functions F. Approximating Integrals

About this book :-
"Machine Learning: A Probabilistic Perspective" by "Kevin P. Murphy" provides a thorough introduction to "probabilistic machine learning", combining theory, algorithms, and practical examples. The book frames machine learning as a problem of "probabilistic modeling", teaching readers how to handle uncertainty and make principled predictions. It covers foundational topics such as "regression", classification, and supervised learning, providing a strong basis for understanding complex models. The text delves into advanced topics, including "Bayesian methods", graphical models, and approximate inference techniques. Murphy explains how these methods allow machines to model uncertainty, reason about data, and improve predictive performance. Real-world examples from computer vision, natural language processing, and time-series analysis show how probabilistic thinking translates into actionable solutions. The book emphasizes both the mathematics behind algorithms and their practical implementation, helping readers bridge theory and practice. Designed for graduate students, researchers, and experienced practitioners, this book equips readers to design robust, interpretable machine learning models and apply them to real-world challenges. By mastering probabilistic approaches, readers gain a deeper understanding of uncertainty, model evaluation, and predictive performance, making it an essential reference for anyone serious about "graphical models", "inference", and probabilistic machine learning.

Book Detail :-
Title: Machine Learning: A Probabilistic Perspective by Kevin Murphy
Publisher: The MIT Press
Year: 2012
Pages: 343
Type: PDF
Language: English
ISBN-10 #: 0262018020
ISBN-13 #: 978-0262018029
License: CC BY-NC-ND 4.0
Amazon: Amazon

About Author :-
The author Kevin Patrick Murphy is a leading "machine learning" and "Bayesian modeling" researcher. Born in Ireland and raised in England, he studied at the "University of Cambridge" (BA), "University of Pennsylvania" (MEng), and earned his Ph.D. in Computer Science from "UC?Berkeley". He completed a postdoctoral fellowship at MIT and published numerous influential papers on probabilistic methods, graphical models, and statistical approaches to "artificial intelligence". Murphy is now a senior "research scientist" at Google. His work focuses on deep probabilistic models, reinforcement learning, and decision-making under uncertainty, helping researchers and practitioners build robust, interpretable "models" for real-world applications.

Similar Regression & Statistical Learning Books