本书以信息选择为视角，全面介绍了稀疏性在机器学习中的应用。通过系统性的讲解，读者能够深入理解信息选择对机器学习算法性能的重要性以及基于稀疏性的信息选择方法。

CHAPTER 1 INTRODUCTION
1.1 Feature Selection
1.2 Transfer Learning
1.3 Outline
CHAPTER 2 BACKGROUND
2.1 Notations
2.2 Single-Layer Autoencoder
2.3 Long-Short-Term Memory Network
2.4 Sparse Learning-Based Unsupervised Feature Selection
2.5 Self-Taught Learning
2.6 Few-Shot Learning
2.7 Hyperspectral Signal Analysis
2.8 Human Activity Recognition
CHAPTER 3 FEATURE SELECTION
3.1 Vertical Federated Learning-Based Supervised Feature Selection
3.2 Supervised Hyperspectral Band Selection
3.3 Unsupervised Feature Selection with Data Structure Preservation
CHAPTER 4 TRANSFER LEARNING
4.1 Graph and Autoencoder-Based Self-Taught Learning
4.2 Few-Shot Learning-Based Cross-Domain Human Activity Recognition
BIBLIOGRAPHY