Every class of the August–November 2022 offering, numbered in order, with its slides, notes and notebooks. Labs were hands-on sessions; their notebooks open in Colab.
Introduction to probabilistic ML1–6 Approximate inference7–12 Bayesian neural networks13–14 Gaussian processes15–16 Unsupervised learning17–18 Active learning and Bayesian optimisation19–20 Autoencoders, hierarchical and generative models21–23 Additional material
Introduction to probabilistic ML Lectures 1–6
Lecture 01Bayes rule and course logistics Lecture 02Lab: distributions in TFP and coin-toss MLE Lecture 03Linear regression: maximum likelihood Lecture 04Logistic regression MLE; coin toss and linear regression MAP Lecture 05Maximum a posteriori estimation Lecture 06Lab: Bayesian linear regression in closed form
Approximate inference Lectures 7–12
Lecture 07Laplace approximation Lecture 08Monte Carlo sampling Lecture 09Monte Carlo sampling, continued Quiz 1 was held in this class.
Lecture 10Rejection sampling Lecture 11MCMC I: Metropolis–Hastings Lecture 12Variational inference
Bayesian neural networks Lectures 13–14
Lecture 13MC dropout and deep ensembles Lecture 14Variational inference and MCMC for neural networks
Gaussian processes Lectures 15–16
Lecture 15Gaussian processes Lecture 16Gaussian processes in code
Unsupervised learning Lectures 17–18
Lecture 17Gaussian mixture models Lecture 18Probabilistic PCA
Active learning and Bayesian optimisation Lectures 19–20
Lecture 19Active learning Lecture 20Bayesian optimisation
Autoencoders, hierarchical and generative models Lectures 21–23
Lecture 21Autoencoders Lecture 22Hierarchical modelling Lecture 23Diffusion models
Additional material 1 topic
· Further slides and notebooks Course files not tied to a single class.
Classes without links had no material posted on the course site.