Module Leader

Professor Bob Shorten

The world is now full of examples of applications in which machine learning algorithms such as image recognition have transformed our lives. The objective of this module is to de-mystify the topic, and expose students, in an accessible manner, to both the basics of machine learning, and to some of its most important methods.       


Learning Outcomes

On completion of this module, students will be better able to: 
  • Employ mathematical tools that underpin machine learning.
  • Apply machine learning algorithms in a lab setting to a range of data sets and problems.
  • Analyse the suitability of a machine learning approach to a problem.
  • Explain the operation and structure of a range of machine learning techniques.

Description of Content

The first part of this module reviews some basic concepts and some elementary mathematics. The second part is then organised around several tutorials (labs and lectures) devoted to the topics of regression and classification, using mainstream machine learning methods. The third part is organised in a similar manner, but focussing on classification from the perspective of Markovian methods and graph theory. Finally, the fourth part gives a birds-eye view of some frontier topics in the field. By the end of the module, students will be familiar with the basics of machine learning, have experienced some of its popular methods, and applied some of its most famous algorithms in a lab setting.

 Fundamentals of Machine Learning:

  • Basic concepts
  • Mathematical tools

Supervised learning algorithms:

  • Regression
  • Classification

Reinforcement learning:

  • Markovian methods

Spectral graph theory;

Frontier topics