Module: Linear Regression

Maximum Likelihood (Ordinary Least Squares) and Bayesian formulations of Linear Regression, geometry and significance

Dates: Wed, Oct 25 - Mon, Nov 6

Learning Outcomes

Formulate linear least squares from Maximum Likelihood and Bayesian perspectives
Understand Bayesian linear least squares and ridge regression
Connect Ridge Regression with conditional means of multivariate Gaussians
Appreciate the distinction between geometry of regression and generalization
Some idea of significance of features ((t-)statistics)
Feature selection using LASSO

Class notes

Class notes

Class notes

Class notes

Class notes

15 Jul 11:00 PM

Geometry -> Generalization

15 Jul 11:00 PM

t- values

15 Jul 11:00 PM

Outcome(s) assessed:

16 Jul 09:00 AM