Machine Learning: Probabilistic Perspective

Machine learning is introduced comprehensively in this text, using probabilistic models and inference as a unifying approach. In today's digital age, there is an overwhelming amount of data available online that requires automated analysis methods. Machine learning provides the tools needed to automatically identify patterns in data and use these patterns to predict future trends.

This textbook offers a thorough and self-contained introduction to the field of machine learning, with a focus on a unified, probabilistic approach. The coverage includes foundational material on topics like probability, optimization, and linear algebra, as well as discussions on recent advancements in the field such as conditional random fields, L1 regularization, and deep learning.

Written in an accessible and informal style, the book includes pseudo-code for important algorithms and is richly illustrated with color images and examples from various application domains such as biology, text processing, computer vision, and robotics. Rather than offering a collection of heuristic methods, the book emphasizes a principled model-based approach, often using graphical models to succinctly and intuitively specify models.

The majority of models discussed in the text are available in a MATLAB software package called PMTK (probabilistic modeling toolkit), which can be accessed online for free. This book is suitable for upper-level undergraduates with a basic understanding of college-level math, as well as graduate students who are just starting out in the field of machine learning.

product information: