2019/20 Undergraduate Module Catalogue
COMP3611 Machine Learning
10 creditsClass Size: 240
Module manager: Matteo Leonetti
Taught: Semester 1 View Timetable
Year running 2019/20
This module is not approved as a discovery module
ObjectivesOn completion of this module, students should be able to:
• list the principal algorithms used in machine learning, and derive their update rules
• appreciate the capabilities and limitations of current approaches;
• evaluate the performance of machine learning algorithms;
• use existing implementation(s) of machine learning algorithms to explore data sets and build models.
Topics selected from:
Neural networks, decision trees, support vector machines, Bayesian learning, instance-based learning, linear regression, clustering, reinforcement learning, deep learning.
Methods for evaluating performance.
Examples will be drawn from simple problems that arise in studies of robotics and computer vision.
|Delivery type||Number||Length hours||Student hours|
|Private study hours||47.00|
|Total Contact hours||53.00|
|Total hours (100hr per 10 credits)||100.00|
Methods of assessment
|Assessment type||Notes||% of formal assessment|
|Total percentage (Assessment Coursework)||20.00|
The coursework is on the application of a supervised learning algorithm to an existing, realistic, dataset. Students are asked to design an appropriate model, evaluate its performance, and analyse the effect of the parameters on the results. The coursework is implemented in python, and makes use of state-of-the-art machine learning libraries.
|Exam type||Exam duration||% of formal assessment|
|Standard exam (closed essays, MCQs etc)||2 hr 00 mins||80.00|
|Total percentage (Assessment Exams)||80.00|
Normally resits will be assessed by the same methodology as the first attempt, unless otherwise stated
Reading listThe reading list is available from the Library website
Last updated: 27/09/2019
Browse Other Catalogues
- Undergraduate module catalogue
- Taught Postgraduate module catalogue
- Undergraduate programme catalogue
- Taught Postgraduate programme catalogue
Errors, omissions, failed links etc should be notified to the Catalogue Team.PROD