0: Preface

These are my notes for CSC311H1: Introduction to Machine Learning at the University of Toronto.

Course Overview

An introduction to methods for automated learning of relationships on the basis of empirical data:

• Classification and regression using nearest neighbour methods, decision trees, linear models, and neural networks
• Clustering algorithms
• Problems of overfitting and assessing accuracy

The Articles

Part I: Supervised Learning — Classification & Regression

1. K-Nearest Neighbors — Distance metrics, decision boundaries, curse of dimensionality
2. Decision Trees — Splitting criteria, information gain, pruning
3. Linear Regression — Least squares, closed-form solution, gradient descent
4. Linear Classification — Logistic regression, softmax, cross-entropy loss
5. Neural Networks — Architecture, activation functions, forward pass
6. Training Neural Networks — Backpropagation, optimization, regularization

Part II: Unsupervised Learning

7. Clustering — K-means, hierarchical clustering, evaluation metrics

Part III: Model Selection & Evaluation

8. Overfitting & Regularization — Bias-variance tradeoff, L1/L2 regularization, dropout
9. Model Evaluation — Cross-validation, precision/recall, ROC curves

Coming Winter 2026.