Lecture 1: Feature Extraction and PCA

In this lecture, we first introduced an important concept in machine learning: feature extraction. Then, we presented Principal Component Analysis (PCA), a linear solution that is essential in learning data-driven methods. After discussing its principles and implementation, we concluded this session with a discussion.

Lecture notes:

• Read the integrated notes: here
• Read the pagination notes: here

Reading guidelines of textbook

Read section 12.2 and Lab 12.5.1

Your tsasks

1. If you are still not familiar with PCA, please carefully read through all of the lecture notes in this session. If you have already learned PCA, use my notes to review and test whether you can understand PCA from the perspective of feature extraction.

2. Familiarize yourself with the R codes in the simple example from Section 3 of the lecture notes. This will be the necessary preparation for the lab session after the second lecture.

Course Homepage