Hide menu

Advanced Bayesian Learning

2014VT

Status Archive
School Computer and Information Science (CIS)
Division STIMA
Owner Mattias Villani
Homepage TBA

  Log in  




Course plan

No of lectures

1-2 lectures (2 x 45 minutes) on each of the six topics.

Recommended for

Anyone with an interest in learning probabilistic models from data.

The course was last given

Never before.

Goals

The aim of the course is to present (approximately) six more advanced topics in Bayesian learning. A preliminary list of topics is (lecturer in parenthesis):

- Prediction and Decision (Mattias Villani)
- Gaussian Processes (Mattias Villani)
- Regularization (Mattias Villani)
- Finite and Infinite Mixture Models (Mattias Villani)
- Bayesian Networks (Patrik Waldmann)
- Sequential Monte Carlo (Thomas Schön)

Prerequisites

The course Bayesian Learning, 6 credits, or a similar course, is a recommended prerequisite.

Organization

Lectures and computer labs.

Contents

- Prediction and Decision (Mattias Villani)
- Gaussian Processes (Mattias Villani)
- Regularization (Mattias Villani)
- Finite and Infinite Mixture Models (Mattias Villani)
- Bayesian Networks (Patrik Waldmann)
- Sequential Monte Carlo (Thomas Schön)

Literature

Scientific articles and individual book chapters.

Lecturers

Mattias Villani (main lecturer)
Patrik Waldmann (Bayesian networks)
Thomas Schön (Sequential Monte Carlo)

Examiner

Mattias Villani

Examination

Computer lab reports.
Individual project.

Credit

6 credits

Comments


Page responsible: Director of Graduate Studies