Introduction to Machine Learning (Fall 2020) (M-I-T)

• Introduction to ML - perspective and history (M-I-T)
• Linear classifiers (M-I-T)
• The random linear classifier algorithm (M-I-T)
• Machine learning as optimization - framework (M-I-T)
• Logistic regression - setting and sigmoid function (M-I-T)
• Linear logistic classifier - hypothesis class (M-I-T)
• Linear logistic classifier - negative log likelihood loss function (M-I-T)
• Machine learning as optimization - gradient descent in one dimension (M-I-T)
• Machine learning as optimization - gradient descent in multiple dimensions (M-I-T)
• One-dimensional linear regression - demo (M-I-T)
• Two-dimensional linear regression - demo (M-I-T)
• Linear logistic classifier - a few comments about regularization (M-I-T)
• Gradient descent optimization - algorithm in one dimension (M-I-T)
• Gradient descent optimization - local optima (M-I-T)
• Gradient descent optimization - algorithm in multiple dimensions (M-I-T)
• Gradient descent optimization - parameters and demo (M-I-T)
• Regression and the ordinary least squares problem (M-I-T)
• Regression - ordinary least squares solution using optimization (M-I-T)
• Regression - OLS analytical solution setup (M-I-T)
• Regression - OLS analytical solution using gradients (M-I-T)
• Regression - beauty of the closed form OLS solution (M-I-T)
• Regression - regularization by ridge regression (M-I-T)
• Regression - analytical minimization of the ridge regression objective (M-I-T)
• Regression - ridge regression using gradient descent (M-I-T)
• Regression - stochastic gradient descent (M-I-T)