| 1 | Machine Learning General Concepts | The vocabulary: model, loss, empirical risk, likelihood, regularization and generalization. |
| 2 | Linear Regression: Modeling and Gradient Descent | Fit a line, compute residuals and follow gradient-descent updates. |
| 3 | Ordinary Least Squares and Regularization | The closed-form least-squares solution, and Ridge, Lasso and Elastic Net. |
| 4 | Softmax Regression | Multiclass classification with logits, softmax probabilities and cross-entropy. |
| 5 | Linear and Quadratic Discriminant Analysis | Gaussian classifiers whose covariance assumption gives a linear or quadratic boundary. |
| 6 | Training a Model: Gradients, Optimizers, and Learning Rates | Follow one parameter update, then minibatches, optimizer state, learning rates and checkpoints. |
| 7 | Decision Trees, Random Forests, and Gradient Boosting | Derive a tree split and a boosting update, then compare their biases and validation needs. |
| 8 | Clustering and Dimensionality Reduction | Find groups with k-means and compress features with PCA. |
| 9 | Reinforcement Learning: Rewards, Value, and Sequential Decisions | Rewards, value and exploration on a two-state example. |
| 10 | Multilayer Perceptron | Tensor shapes, backpropagation and a worked XOR network. |
| 11 | Activations and Gated Feed-Forward Networks | ReLU, GELU, SiLU and the GLU variants used in modern feed-forward layers. |
| 12 | Convolutional Neural Networks | Convolution, channels, padding, stride and pooling; equivariance versus invariance. |
| 13 | Recurrent Neural Networks | Recurrent state, backpropagation through time, LSTM gates and state-space models. |
| 14 | Reading “Attention Is All You Need”: The Original Transformer Paper | The 2017 paper read closely: architecture, training setup and results. |
| 15 | Variational Autoencoders | Latent-variable generation with the ELBO and the reparameterization trick. |
| 16 | Diffusion Models | Forward noising, learned denoising and iterative sampling. |
| 17 | Datasets: Provenance, Leakage, and Evaluation | Where a dataset comes from, and what its collection process lets you claim. |
| 18 | Model Evaluation Under Distribution Shift | Evaluate for the population you will deploy on, not for one test set. |