Mathematics for Machine Learning

Marc Peter Deisenroth, A. Aldo Faisal, Cheng Song Ong

Digital versions PDF
Open source No
Exercises Yes
Solutions Yes
Instructor’s manual Available upon request
Additional exercises with solutions Yes
Print version Available from Cambridge University Press
License Copyright 2021 by authors, all rights reserved
  • Text to support a machine learning course summarizing the necessary mathematical background
  • PDF version freely available: 12 chapters, 405 pages
  • A reasonably priced print version is available
  • A limited set of homework exercises is supplemented by an online repository of additional exercises with solutions
  • An instructor’s manual is available upon request of the publisher
  • Tutorials are provided in accompanying Jupyter notebooks
  • For more information and to download

This textbook is meant to summarize the mathematical underpinnings of important machine learning applications and to connect the mathematical topics to their use in machine learning problems. The authors state, “The book is not intended to cover advanced machine learning techniques because there are already plenty of books doing this. Instead, we aim to provide the necessary mathematical skills to read those other books.” Some readers may find the mathematical exposition to be somewhat terse as the authors intend to describe results that will be needed later rather than building the mathematical theory from the ground up. Numerous examples convincingly illustrate how the mathematical topics are used in machine learning applications. The text could also be used to supplement, say, a second linear algebra course with current applications.

Table of Contents

  1. Part 1. Mathematical Foundations
    1. Introduction and Motivation
    2. Linear Algebra
    3. Analytic Geometry
    4. Matrix Decompositions
    5. Vector Calculus
    6. Probability and Distribution
    7. Continuous Optimization
  2. Part II: Central Machine Learning Problems
    1. When Models Meet Data
    2. Linear Regression
    3. Dimensionality Reduction with Principal Component Analysis
    4. Density Estimation with Gaussian Mixture Models
    5. Classification with Support Vector Machines