COMP47590 Advanced Machine Learning

Academic Year 2024/2025

COMP47590 is an advanced module on Machine Learning that builds on the core concepts covered in COMP47490 or COMP47460. Either COMP47490 or COMP47460 is a prerequisite for this module. This module covers advanced, state of the art topics in machine learning in areas such as deep learning, ensemble methods, semi-supervised learning, human-in-the-loop machine learning, unsupervised machine learning, and reinforcement learning. Significant prior programming experience is essential (ideally in Python (with associated machine learning libraries), but experience in Java, C/C++, etc and a willingness to learn Python independently would suffice).

Show/hide contentOpenClose All

Curricular information is subject to change

Learning Outcomes:

On completion of this module, students will be able to: 1) Distinguish between the different categories of machine learning algorithms; 2) Identify suitable machine learning approaches for different tasks, including state-of-the-art methods; 3) Run and evaluate a range of algorithms for different tasks using a standard machine learning toolkit; 4) Implement and evaluate machine learning algorithms in a high-level language.

Student Effort Hours: 
Student Effort Type Hours
Autonomous Student Learning

72
Lectures

24
Practical

8
Total

104
Approaches to Teaching and Learning:
The main teaching and learning approaches used in this module are lectures, lab work, task-based learning, group work and student-led readings. 
Requirements, Exclusions and Recommendations
Learning Requirements:

Having previously completed an introductory machine learning module (such as COMP47490, COMP47460, or COMP47750) is a prerequisite for this module.
To complete the continuous assessment, this module requires significant prior programming experience, ideally in Python.

Learning Recommendations:

Students should have strong mathematical ability, as some of the algorithms require some understanding of linear algebra and statistical concepts.


Module Requisites and Incompatibles
Incompatibles:
COMP47650 - Deep Learning


 
Assessment Strategy  
Description Timing Open Book Exam Component Scale Must Pass Component % of Final Grade
Exam (In-person): End of year final exam. n/a Alternative linear conversion grade scale 40% No

40
Individual Project: Project involving development of machine learning solutions. May be completed in small groups or individually. n/a Alternative linear conversion grade scale 40% No

26
Participation in Learning Activities: Participation in ungraded, but mandatory, in-class development, reading group, and other activities. n/a Alternative linear conversion grade scale 40% No

8

Carry forward of passed components
No
 
Resit In Terminal Exam
Summer No
Please see Student Jargon Buster for more information about remediation types and timing. 
Feedback Strategy/Strategies

• Feedback individually to students, post-assessment
• Group/class feedback, post-assessment
• Online automated feedback

How will my Feedback be Delivered?

feedback will be provided following assignment submission and correction.