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

Machine learning as optimization – gradient descent in one dimension (M-I-T)

Model-based learning (M-I-T)

Nearest neighbor models (M-I-T)

Neural networks – activation functions (M-I-T)

Neural networks – basic element (M-I-T)

Neural networks – brief review of layers and backprop (M-I-T)

Neural networks – layer definition (M-I-T)

Perceptron through origin algorithm (M-I-T)

Neural networks – many layers (M-I-T)

Policy search (M-I-T)

Neural networks – optimizing parameters – adagrad and adadelta (M-I-T)

Proof sketch of the perceptron convergence theorem (M-I-T)

Neural networks – optimizing parameters – adam step-size update strategy (M-I-T)

Q-learning (M-I-T)

Neural networks – optimizing parameters – adaptive step-size (M-I-T)

Neural networks – optimizing parameters – batch gradient descent training (M-I-T)

Q-learning select-action strategies (M-I-T)

Neural networks – optimizing parameters – momentum (M-I-T)

Random forests models (M-I-T)

Recommender systems – introduction (M-I-T)

Neural networks – optimizing parameters – running averages (M-I-T)

Neural networks – output layer activation functions (M-I-T)

Recurrent neural network model (M-I-T)

Neural networks – regularization by batch normalization (M-I-T)

Regression – analytical minimization of the ridge regression objective (M-I-T)