MLPR 2023 Activities Notes Forum FAQ Feedback Accessibility

MLPR class notes

When learning advanced material, you won't immediately understand everything just from reading notes. Please sign up to the forum, ask questions, and share insights and external materials that you have discovered.

Lecture recordings and scans are on the Activities page.

Keyboard shortcut: step through the notes using the left and right arrow keys.

Please annotate the HTML versions of the notes in the forum, to keep the class's comments together. You can show/hide links to PDF versions for printing. However, the PDFs don't have the in-note questions.

Background information

Week 1: Introduction to ML with Linear Regression

Week 2: ML fundamentals: generalization, error bars, Gaussians

Week 3: Classification and gradient-based fitting

Week 4: Bayesian linear regression

Week 5: Bayesian model choice and Gaussian processes

Week 6: More detailed models: Gaussian process kernels, more non-Gaussian regression

Week 7: Neural Networks

Week 8: Autoencoders, PCA, Netflix Prize

Week 9: Bayesian logistic regression, Laplace approximation

Week 10: Sampling-based approximate Bayesian inference, variational inference

Status Page: Check which in-note questions you might have missed.

That’s the end of the course. Information about the exam is on the activities timeline. The course survey is open until 21 December.

 

A coarse overview of major topics covered is below. Some principles aren't taught alone as they're useful in multiple contexts, such as gradient-based optimization, different regularization methods, ethics, and practical choices such as feature engineering or numerical implementation.

You are encouraged to write your own outlines and summaries of the course. Aim to make connections between topics, and imagine trying to explain to someone else what the main concepts of the course are.