Lecture
This lecture covers the concept of linear least squares, focusing on finding coefficients that minimize the error in a linear model. It explores normal equations, characterisation of solutions, and the gradient descent method. The lecture also delves into the importance of linear regression and the process of minimizing squared errors. Additionally, it discusses the uniqueness of solutions and the use of matrix form to represent the linear model. The lecture concludes with a detailed explanation of simple linear regression and the application of Gaussian kernel functions.