intensive reading:
- 1 Introduction
- 2 model evaluation and selection
- 3 linear model
- 4 Decision Tree
- 5 Neural Networks
- Cluster 9
Crude read:
- 6 SVM
- 7 Bayesian classifier
- 8 Integrated Learning
- 10 metric learning and dimensionality reduction
- 11 feature selection and sparse learning
- 12 calculated Learning Theory
unread:
- 13 semi-supervised learning
- 14 probabilistic graphical models
- 15 rule learning
- 16 Reinforcement Learning
This book features are: comprehensiveness, the contents of a lot of machine learning are described, and the last one when there are problems encountered chapter of recommended papers.
But not very detailed. Embodied in the certification process is simple, requires a considerable degree of foundation.
This book does not read all. In the next step plan to combat specific learning, and have problems in the future do to add
Combination of these two features, the book uses:
- The concept can be used as a book of their own learning machine, so a lot of content on your own machine learning an idea,
- When you are experiencing problems, you can find a way to solve the problem corresponding paper in this book. To be sure, this book just tells
Any way, but needs its own online search more specific detailed tutorial
- When his own machine learning have a certain understanding, you can use this book to check where their master is not good enough, the so-called recovery disk.