Lecture
This lecture starts with a recap on parametric models, hyperplanes, linear regression, and multi-output prediction. It then delves into binary classification using linear regression, the least-square approach, and the logistic function. The drawbacks of the step function and the need for non-linearity are discussed. The lecture covers gradient-based optimization, minimizing functions, and the application of gradient descent. It concludes with logistic regression training, gradient computation, and model evaluation metrics like accuracy, precision, and recall.