Machine Learning and Covariance Matrices
摘要
This chapter first reviews the three types of machine learningMachine learning: supervised learningSupervised learning, unsupervised learningUnsupervised learning, and semi-supervised learningSupervised learningSemi-supervised learning, and briefly discusses reinforcement learningReinforcement learning. Subsequently, we introduce three types of deep learningDeep learning methods: Convolution Neural Networks (CNNs)Convolutional neural networks (CNNs), Graph Convolutional Networks (GCNs),Graph convolutional networks (GCNs) and Transformers, and we discuss their relationships with covariance matrices. We then introduce the concept of transfer learningTransfer learning and its usefulness in high-dimensional covariance analysis. An empirical example is presented to illustrate the application of deep learningDeep learning.