Lecture 5: Regression Models
In this lecture, we discuss regression problems. First, we review the training methods of linear regression models from different perspectives. Then, a basic nonlinear extension idea is introduced, feature mapping, and use it to introduce the first nonlinear regression model, polynomial regression. Afterward, we understand polynomial regression from the perspective of basis functions, and then introduce another commonly used nonlinear model, spline regression. At the end of this lecture, we introduce a new concept, the overfitting problem. It is a core challenge in machine learning from the training perspective, and we focus on it in the next lecture.
Outline:
- 5.1 Linear Regression Model
- 5.2 Nonlinear Regression Model
- 5.3 Ovrfitting Problem
