CMT311: Principles of Machine Learning
School | Cardiff School of Computer Science and Informatics |
Department Code | COMSC |
Module Code | CMT311 |
External Subject Code | 100992 |
Number of Credits | 20 |
Level | L7 |
Language of Delivery | English |
Module Leader | Dr Yazmin Ibanez Garcia |
Semester | Autumn Semester |
Academic Year | 2025/6 |
Outline Description of Module
Machine Learning (ML) is a subfield of AI that studies methods for developing computer programs that learn from examples or from prior experience. This module provides an introduction to foundational aspects of the field. The core idea of ML is to study learning processes, including designing algorithms that can predict outcomes, or extract knowledge from data. The primary objective of this module is to introduce the concepts of machine learning, including various algorithms, underlying principles, and theoretical foundations.
On completion of the module a student should be able to
-
Discuss the motivation behind common machine learning approaches.
-
Describe concepts and algorithms widely used in machine learning.
-
Formulate the theoretical principles underpinning machine learning methods and algorithms.
-
Apply machine learning algorithms to simple examples.
-
Compare different approaches and algorithms for machine learning.
-
Determine a suitable machine learning approach given an application.
How the module will be delivered
Delivery of this module will encourage active learning. You will be guided through learning activities targeting the learning outcomes of the module.
Theoretical material delivered via pre-recoded video-lectures and reading material, to be revised by students prior to the weekly sessions (e.g. lectures and tutorials).
Weekly in-person sessions will include activities expanding upon the concepts introduced in the video-lectures and reading assignments.
Each session will consist of
-
a discussion portion where students discuss the materials that they had studied, and
-
a practical exercise portion where students work in small groups on an exercise applying the concepts from lecture/readings to a small example.
Skills that will be practised and developed
Abstract thinking and modelling
Understanding of expressivity – efficiency trade-off in AI
Ability to understand mathematical theories, their assumptions and consequences
How the module will be assessed
There are two assessments during the semester, each worth 50% of the module marks.
The first assessment is coursework and will test learning outcomes LO4, LO5, and LO6.
The second assessment is an exam and will test learning outcomes LO1, LO2, and LO3.
Students will be provided with reassessment opportunities in line with University regulations.
Assessment Breakdown
Type | % | Title | Duration(hrs) |
---|---|---|---|
Written Assessment | 50 | Machine Learning Methods | N/A |
Exam online – Autumn semester | 50 | Theoretical Foundations Of Machine Learning | 1.5 |
Syllabus content
-
Basics of (statistical) Machine Learning
-
Inference and prediction
-
Selected classification and regression models
-
Basics of Machine Learning theory: PAC-learning, VC Dimension