Hide menu

732A46 Bayesian Learning

Timetable


The basics, single- and multiparameter models
What? When? Where? Read? Contents Material
Lecture 1 Oct 27, 10-12 VN BDA, Ch. 1-2. Likelihood. Intro to Bayesian inference. Bernoulli model.
Lecture 2 Oct 28, 10-12 TBA BDA, Ch. 1-2. Normal and Poisson models. Conjugate and non-informative priors.
Lecture 3 Oct 28, 13-15 TBA BDA, Ch. 3.1-3.3 and
3.5-3.6
Multiparameter models, Marginalization, Normal with unknown variance, Multinomial, Multivariate normal
Computer 1 Oct 30, 10-12 PC4-5 Gridding posteriors. Basic simulation.
Regression models
What? When? Where? Read? Contents Material
Lecture 4 Nov 4, 10-12 TBA Prediction and making decisions
Lecture 5 Nov 6, 10-12 TBA BDA 4.1-4.2, 14.1-3 and 14.6-7. Linear regression. Binary regression. Posterior approximation.
Lecture 6 Nov 11, 10-12 VN Chapter 2.1-2.3 and 2.5 in this book Non-linear regression. Shrinkage priors. Gaussian processes.
Computer 2 Nov 12, 10-12 PC4-5 Bayesian polynomial regression with conjugate prior.
Tackling more advanced models with MCMC
What? When? Where? Read? Contents Material
Lecture 7 Nov 19, 10-12 VN BDA Ch. 10-11 Gibbs sampling and data augmentation
Lecture 8 Nov 20, 10-12 VN BDA Ch. 10-11 Markov Chain Monte Carlo, Metropolis-Hastings
Lecture 9 Nov 21, 15-17 TBA BDA Ch. 10-11 Hierarchical models
Computer 3 Nov 25, 10-12 PC1-2 Gibbs sampling and binary regression.
Model Inference
What? When? Where? Read? Contents Material
Lecture 10 Dec 1, 10-12 TBA Article on Bayes factors Model inference
Lecture 11 Dec 1, 13-15 VN Model inference
Lecture 12 Dec 2, 10-12 TBA Bayesian variable selection.
Computer 4 Dec 4, 10-12 PC4-5 The Metropolis algorithm
Project
Project deadline Dec 21, Midnight.
Note: Linköping University makes use of so called academic quarters.
BDA = Bayesian Data Analysis book
VN = Von Neumann
TBA = To be announced.

Page responsible: Mattias Villani
Last updated: 2014-10-20