Probabilistic Machine Learning: An Introduction, Kevin Murphy

PDF, Exercices solutions
In this book, we will cover the most common types of ML, but from a probabilistic perspective. Roughly speaking, this means that we treat all unknown quantities (tomorrow’s temperature, or the parameters of some model) as random variables, that are endowed with probability distributions that describe a weighted set of possible values the variable may have. There are two main reasons we adopt a probabilistic approach. First, it is the optimal approach to decision-making under uncertainty. Second, probabilistic modeling is the language used by most other areas of science and engineering and thus provides a unifying framework between these fields.

I. Foundation

3. Probability: Multivariate Models
4. Statistics
5. Decision theory
7. Linear Algebra
8. Optimization

II. Linear Models

9. Linear Discriminant Analysis
10. Logistic Regression
11. Linear Regression
12. Generalized Linear Models

III. Deep Neural Networks

13. Neural networks for tabular data
14. Neural networks for images
15. Neural networks for sequences

IV. Non parametric model

16. Exemplar-based Models
17. Kernel Methods
18. Trees, Forests, Bagging and Boosting

V. Beyond supervised learning

19. Learning with Fewer Labeled Examples
20. Dimensionality reduction
21. Clustering
22. Recommender Systems