# Statistical Machine Learning (SML) PhD course (9+3hp)

### Note that the 2018 MSc level course is available here.

### The 2018 edition of the PhD course is on another page

#### Spring 2016

### Description

Data is becoming more and more widely available and the world is now in a situation where there is more data than we can handle. This clearly calls for new technology and this challenge has resulted in the rapid growth of the machine learning area over the past decade. This course provides an introduction into the area of machine learning, focusing on dynamical systems. To a large extent this involves probabilistic modeling in order to be able to solve a wide range of problems.

### Contents

- Linear regression
- Linear classification
- Support vector machines
- Expectation Maximization (EM)
- Neural networks
- Clustering
- Approximate inference (VB and EP)
- Graphical models
- Message passing algorithms
- Sampling methods and MCMC
- Bayesian nonparametric (BNP) models

### Course Structure

The course gives 9 hp (you can receive an additional 3 hp by carrying out a project).

- Lectures: 11
- Problem solving sessions: Coordinated by Fredrik Olsson, a schedule is available via TimeEdit by clicking here.
- Project: Optional

### Examination

The examination consists in a standard written 3 day (72 h) exam. The exam period is 31/3-2016 until 29/4-2016.

### Course literature

The main book used during the course is,

[B] Christopher M. Bishop. Pattern Recognition and Machine Learning, Springer, 2006.

We will also make use of,

[HTF] Trevor Hastie, Robert Tibshirani and Jerome Friedman. The Elements of Statistical Learning: Data Mining, Inference and Prediction, Second edition, Springer, 2009.

### Recommended supplementary reading

There are by now many books written on the machine learning subject and new books keeps appearing all the time. Here are links to a few additional resources.

- Kevin P. Murphy. Machine learning - a probabilistic perspective, MIT Press, 2012.
- Daphne Koller and Nir Friedman. Probabilistic Graphical Models Principles and Techniques, MIT Press, 2009.
- David Barber. | Bayesian Reasoning and Machine Learning, Cambridge University Press, 2012.
- Sergios Theodoridis. Machine Learning, A Bayesian and Optimization Perspective, 2015.
- Gareth James, Daniela Witten, Trevor Hastie and Robert Tibshirani. An Introduction to Statistical Learning - with Applications in R, Springer, 2013. (provides a nice introduction to the area of statistical machine learning for non-mathematical sciences)

### Periodicity

Every 2 years. Next edition starts in Jan. 2016. Previously given at Uppsala University (2014), Linköping University (2011, 2013) and at Lund University (2011).

### Schedule

The course schedule is available via TimeEdit by clicking here.

### Course level

This is a PhD level course.

### Prerequisites

Basic undergraduate courses in linear algebra, statistics, signal and systems.

### Related Courses

Statistical estimation theory and its applications

Foundations in machine learning (very small overlap with this course, hence they complement each other well), Computational learning in dynamical systems.

### Contact Person

Prof. Thomas Schön, email: thomas.schon_at_it.uu.se.