Module information on this degree can be found below, separated by year of study.

The module information below applies for the current academic year. The academic year runs from August to July; the 'current year' switches over at the end of July.

Students select optional courses subject to rules specified in the Mechanical Engineering Student Handbook,  for example at most three Design and Business courses. Please note that numbers are limited on some optional courses and selection criteria will apply.

Machine Learning

Module aims

This module aims to introduce engineering students to Machine Learning (ML). ML is an increasingly high-profile technology, having huge impact across a number of areas. Future engineers are likely to come across ML in some form in their careers; they could be required to solve problems where ML may provide a solution, or may have to assess an existing ML approach which has been developed. This course therefore aims to provide today's engineering graduates with an understanding of ML techniques, including being able to critically assess whether or not ML is suitable for a problem, as well as what type of algorithm would be best and what may be required to implement it. 

ECTS = 5

Learning outcomes

At the end of the module, the student should be able to:

1. Select a suitable machine learning approach (model and data collection) for a particular problem, including assessing whether machine learning is the best solution at all

2. Write Python code to implement a basic demonstration ML algorithm

3. Investigate the performance of different machine learning algorithms for engineering problems

4. Explain the concepts of machine learning techinques and the corresponding terminology

Module syllabus

Bayesian decision theory and maximum likelihood.

Regression approaches.

Linear discriminant functions.

Support vector machines.

Multilayer neural networks.

Nonparametric techniques.

Nonmetric methods.

Unsupervised learning and clustering.

Teaching methods

Lectures will provide students with understanding of underlying techniques within different machine learning methods. Tutorial time will enable students to practice utilising machine learning libraries and tools in order to develop experience. 

Assessments

Assessment details        
      Pass mark   
Grading method Numeric   50%
         
         
Assessments        
Assessment type Assessment description Weighting Pass mark Must pass?
Coursework Project 40% 50% N
Examination Exam 60% 50% N

Reading list

Resources

Module leaders

Dr Peter Huthwaite