CMPUT 467/567

Machine Learning II, Winter 2026

Readings

Students are expected to read the corresponding sections about a class’s topic from notes before class as each class will discuss each topic in more detail and address questions about the material.

All readings are from the (in progress) machine learning notes. These are designed to be short, so that you can read every chapter. I recommend avoiding printing these notes, since later parts of the notes are likely to be modified.

Schedule

This schedule may have small changes throughout the semester. Most of the lectures will be whiteboard lectures. The links to pdf files will be dead links, until we get to that lecture. I will upload all slides right before lecture, so you can follow along during class. The whiteboard lecture notes will be uploaded to a shared Google drive of lecture notes.

Week Date Topic Readings
1 January 6, 2026 Introduction to the Course with a brief refresher of the basics of ML taught in ML I Assignment #1 released

Read Chapters 1-5 from the ml2-notes.pdf
1 January 8, 2026 Probability Concepts (multivariate distributions, mixture models, KL divergences), Probability slides and The Goal of Data representations and Separability slides  
2 January 13, 2026 Revisiting Linear Regression, with matrices, including generalization error and bias and variance (move to whiteboard, lecture notes in this folder)  
2 January 15, 2026 Optimization concepts (Hessians, stepsize algorithms, minibatch sizes, convergence rates) January 16 the Add/Drop deadline (no fees for change)
3 January 20, 2026 Optimization concepts continued Readings Exercises #1 due on Monday, Jan. 19

Readings Exercises #2:
- read Chapters 6-8
3 January 22, 2026 Generalized Linear Models and Multinomial Logistic Regression Assignment #1 due on Friday, Jan. 23

Assignment #2 released
4 January 27, 2026 Constrained Optimization and

First in-class quiz
In-class Quiz 1 about Assignment 1 (must attend class), at start of class
4 January 29, 2026 Cross Validation to estimate generalization error and do hyperparameter selection, slides  
5 February 3, 2026 Learning data representations: latent variable methods and (probabilistic) PCA Readings Exercises #2 due Monday, Feb. 2

Readings Exercises #3:
- read Chapters 9-11
5 February 5, 2026 Learning data representations with neural networks Assignment #2 due on Friday, Feb. 6

Assignment #3 released
6 February 10, 2026 Regularization and inductive biases in neural networks, and leveraging ideas from latent representations and

second in-class quiz
In-class Quiz 2 about Assignment 2
6 February 12, 2026 Finish learning data representations  
7 February 16-20, 2026 No classes, Reading Week  
8 February 24, 2026 Starting with a simple generative model: mixture models Readings Exercises #3 due Monday, Feb 23

Readings Exercises #4:
- read Chapters 12-15
8 February 26, 2026 Generative models with learned representations: variational auto-encoders and the ELBO, and start Midterm Review  
9 March 3, 2026 Finish Midterm Review and do practice midterm Midterm is about topics from Chapter 1-11
9 March 5, 2026 Midterm in-class for the whole 90 minutes  
10 March 10, 2026 Optimizing the ELBO, conditonal VAEs and Evaluating Generative Models  
10 March 12, 2026 Missing Data Assignment #3 due on Friday, March 13

Assignment #4 released
11 March 17, 2026 Uncertainty estimation: Bayesian regression, Gaussian processes and bootstrap ensembles

and third in-class quiz
Readings Exercises #4 due Monday, March 16

Readings Exercises #5:
- read Chapters 16-18
11 March 19, 2026 Finish uncertainty estimation  
12 March 24, 2026 Learning on sequential (temporal) data: history, RNNs and transformers  
12 March 26, 2026 Finish learning on sequential data  
13 March 31, 2026 Generalization error beyond iid  
13 April 2, 2026 Finish topics Assignment #4 due on Friday, April 3
14 April 7, 2026 Fourth In-class quiz 4 and Guest Lecture Readings Exercises #5 due Monday, April 6
14 April 9, 2026 Final Review  
Final April 22, 9 am, in the Main Gym Final Exam 2 hours, Four-page cheat sheet allowed (2 pages front and back)

Link to the schedule from Winter 2025