Explaining Logistic Regression

January 15, 2023

• Different Loss Functions
• Cost Function
• checking the model accuracy
• explaining 2 different solutions using least squares regression and gradient descent
• Using a small example, explain all the mathematics crystal clear

Regularization

$w^Tw=w_0^2+w_1^2+w_2^2+…+w_d^2$ - also known as L2 regularization

It encourages $w_0,…,w_d$ to be small chapter 7 slide 28

Cost Function $argmin\ (Pw - y)^T\ (Pw - y)\ +\lambda w^Tw$