Optional material

While selected material from the resources below may form part of the compulsory reading, the material is generally not required reading. In particular, you will not need to procure any books for the course.

Additional support for the lectures

This is a collection of additional material that may help you to better understand the lectures. Advanced material, sometimes going well beyond the scope of PMR, is marked with a ⚠.

Background

• MLPR math background
• MLPR background on expectation/average
• Koller's video on probability distributions
• Koller's video on statistical independence
• Koller's overview on probabilistic graphical models
• Barber: Chapter 1

Directed graphical models

• Koller's introduction to directed graphical models (Bayesian networks)
• Koller's video on reasoning patterns
• Koller's video on flow of probabilistic influence/introduction to d-separation
• Koller's video on independencies in directed graphical models
• Koller's video on Naive Bayes
• Koller's video on application to medical diagnosis
• Bishop Sections 8.1 and 8.2
• Barber: Chapter 2, Sections 3.1 and 3.3 (without 3.3.6)
• Section 2.1 in Michael Jordan's "An Introduction to Probabilistic Graphical Models"

Undirected graphical models

• Koller's video on factors
• Koller's video on Gibbs distributions
• Koller's video on pairwise Markov networks
• Koller's video on independencies in Markov networks
• Bishop: Section 8.3
• Barber: Section 4.1, 4.2
• Section 2.2 in Michael Jordan's "An Introduction to Probabilistic Graphical Models"

Expressive power and comparison between directed and undirected graphical models

• Koller's video on I-maps and perfect maps
• Barber: Section 4.5
• Chapter 4 of Michael Jordan's "An Introduction to Probabilistic Graphical Models" (file is provided as chapter16.ps) ⚠

Factor graphs

• Bishop: Section 8.4.3
• Barber: Section 4.4
• Research paper "Extending Factor Graphs so as to Unify Directed and Undirected Graphical Models", UAI 2003, by B. Frey ⚠

Exact inference

• MLPR notes on multivariate Gaussians
• Koller's video on (conditional) probability queries
• Koller's video on variable elimination
• Koller's video on the complexity of variable elimination
• Koller's video on graph-based perspective of variable elimination
• Koller's video on finding elimination orderings
• Koller's video on MAP inference
• Bishop: Section 8.4 till 8.4.6 on inference in graphical models
• Barber: Sections 5.1 to 5.4 on variable elimination, sum-product algorithm, etc; Section 23.1 on Markov chains, Section 23.2 on HMM inference
• Research paper "Factor Graphs and the Sum-Product Algorithm", 2001, by F. Kschischang et al. ⚠

Learning, factor analysis, and ICA

• Murphy's notes on conjugate Bayesian analysis of the Gaussian distribution
• Barber: Sections 9.1 to 9.4 on learning; Chapter 21 on factor and independent component analysis
• More details on PCA (Chapter 2) and background on linear algebra (Appendix A)
• ICA demo
• Advanced review on ICA by A. Hyvarinen ⚠

Score matching and other methods to estimate unnormalised models

• Review paper on further estimation methods for unnormalised models by M. Gutmann and A. Hyvarinen
• Original paper on score matching by A. Hyvarinen
• A general framework to estimate unnormalised models by M. Gutmann and J. Hirayama ⚠
• Noise-contrastive estimation to estimate unnormalised models by M. Gutmann and A. Hyvarinen ⚠
• Score matching for generative modelling by Y. Song and S. Ermon ⚠

Sampling and Monte Carlo

• Barber: Chapter 27
• Iain Murray's tutorial on Monte Carlo methods ⚠
• An Introduction to MCMC for Machine Learning by Andrieu et al ⚠
• Introducing Monte Carlo Methods with R by Robert and Casella ⚠
• Pareto Smoothed Importance Sampling by Vehtari et al ⚠ ⚠
• Art Owen's book Monte Carlo theory, methods and examples ⚠ ⚠
• Chopin and Papaspiliopoulos' book An Introduction to Sequential Monte Carlo ⚠ ⚠

Variational inference and EM algorithm

• Barber: Chapter 11
• Shakir Mohamed's tutorial on variational inference
• An Introduction to Variational Autoencoders by Kingma and Welling
• Variational Inference: A Review for Statisticians by Blei et al ⚠
• Advances in Variational Inference by Zhang et al ⚠

Books

This is a list with general and specialised machine learning books. For some of the books, direct links to the library copies are available on the resource list page on Learn. The library’s search portal is here.

• Pattern Recognition and Machine Learning
  Christopher Bishop
  Springer 2006
  (available online)
• Probabilistic Graphical Models : Principles and Techniques
  Daphne Koller and Nir Friedman
  MIT Press 2009
  (available at the University library)
• Bayesian Reasoning and Machine Learning
  David Barber
  Cambridge University Press 2012
  (available online)
• Bayesian networks : an introduction
  Timo Koski and John Noble
  John Wiley 2009
  (available at the University library)
• Graphical models
  Steffen L. Lauritzen
  Oxford University Press 1996
  (available at the University library)
• Probability and Statistics
  Morris DeGroot and Mark Schervish
  Pearson, 4th Edition
  (available at the University library)
• Information Theory, Inference, and Learning Algorithms
  David J.C. MacKay
  Cambridge University Press 2003
  (available online)
• The Matrix Cookbook
  Petersen and Pedersen
  (available online)
• Graphical models and message-passing algorithms: Some introductory lectures
  Martin J. Wainwright
  (available online)
• An Introduction to Probabilistic Graphical Models (incomplete draft)
  Michael Jordan
  (parts available online here and here)
• Independent Component Analysis
  Aapo Hyvarinen, Juha Karhunen, and Erkki Oja
  (electronic copy available at the University library)
• Monte Carlo Statistical Methods
  Christian Robert and George Casella
  Springer 2004
• Introducing Monte Carlo Methods with R
  Christian Robert and George Casella
  Springer 2010
  (available at the University library)
• Bayesian data analysis
  Andrew Gelman, John B. Carlin, Hal S. Stern, David B. Dunson, Aki Vehtari, and Donald B. Rubin
  CRC Press 2013
  (available at the University library)
• Mathematics for Machine Learning
  Marc Peter Deisenroth, A. Aldo Faisal, Cheng Soon Ong
  Cambridge University Press 2020
  (available online)
• An Introduction to Sequential Monte Carlo
  Nicolas Chopin and Omiros Papaspiliopoulos
  Springer 2020
  (available at the University library)
• Art Owen
  Monte Carlo theory, methods and examples
  (available online)
• Probabilistic machine learning book series by Kevin Murphy
  (available online)