Module Catalogue 2024/25

Pre Requisite Comment

Basic knowledge of linear algebra, calculus, and statistics.

Basic knowledge of signals and systems.

Basic skills of computer programming, either C or Matlab or Python or equivalent language.

Knowledge in python is desirable but not necessary

Co Requisite Comment

N/A

Aims

This module aims to provide underlying mathematical, statistical and theoretical concepts of Machine Learning along with essential programming skills and expertise to design, build, and implement appropriate Machine Learning techniques for various engineering applications. The module introduces classical regression, classification, clustering and modern deep learning models. The module includes relevant programming exercises to complement the theoretical concepts, which will allow students to gain valuable conceptual and programming skills to build, optimise and implement these models into a range of practical engineering challenges.

Outline Of Syllabus

This module introduces mathematical foundations and fundamental concepts of machine learning, including supervised, unsupervised, clustering, and modern deep learning methods. The module focuses on mathematically formulating, and practically designing, testing, and validating different types of machine learning algorithms to address various real-world applications. The course delves into data processing and compression techniques, aiding in the interpretation of crucial information extracted by diverse machine learning models. Moreover, the module covers contemporary decision trees, random forests and deep learning algorithms, exploring concepts such as neurons and brains, artificial neural networks, multilayer perceptron, convolutional neural networks, recurrent neural networks, and reinforcement learning. Additionally, it provides an introduction to emerging machine learning paradigms like Tsetlin machines. Ethical considerations, as well as issues regarding privacy and security related to data-driven machine learning models, are also addressed in this module.

Intended Knowledge Outcomes

By the end of the course:
1.       Students will be able to explain the mathematical foundations of machine learning including linear algebra, statistics and probability, and calculus. M1
2.       Students will be able to define fundamental concepts of supervised, unsupervised, clustering, and deep learning methods. M1
3.       The students will be able to select, design and implement the correct machine learning algorithm by distinguishing whether the given problem is a regression, classification, or clustering problem. M2, M3
4.       Students will be able to examine and interpret key information extracted by different machine learning models. M2
5.       Students will be able to explain machine learning pipelines for modern engineering applications. M1
6.       Students will be able to judge ethical and societal issues related to fairness, privacy, and security of data-driven machine learning models. M8

Intended Skill Outcomes

By the end of course, students will be able to:

1.       Algorithmically formulate and design machine Learning models to employ them in several practical applications. M2, M3
2.       Design, debug, and interpret code to train, test, optimise and validate the built models and evaluate their performance. M3
3.       Process, compress, and prepare data to derive and distinguish key structures and information. M2, M3
4.       Engineer a low-complexity machine learning pipeline for real-world applications and domain specific problems. M3
5.       Identify and appraise data biases and decision biases related to data-driven machine Learning models. M1

Teaching Rationale And Relationship

The module employs blended teaching methods, including recorded lecture videos, scheduled in-person lectures, and Q&A/discussion sessions, which encompass a diverse array of teaching approaches for delivering the theoretical and conceptual foundations of machine learning. Additionally, problem-based programming lab sessions provide opportunities to construct, test, and implement the machine learning concepts and theories covered in the lectures.

Assessment Rationale And Relationship

The final NUMBAS-based exam will evaluate students' knowledge and understanding on the fundamental principles and applications of machine Learning techniques. Additionally, the exam will assess students' abilities to formulate, design, and select various machine learning models, as well as their capacity to interpret key information derived from these models. Furthermore, it will test students' capability in evaluating ethical and societal issues associated with data-driven machine learning models.

The formative machine learning programming exercises will reinforce understanding of the machine learning methods and their practical implementation/testing. The formative online quizzes will provide feedback to students on their grasp of these topics and readiness for the exam.

General Notes

Anticipating an enrolment of over 200 students for the module in AY 2024-25, ensuring personalised learning experiences and addressing individual needs necessitates additional staff support in the form of extra lecturers and PGR demonstrators. Therefore, the module leadership team intends to request supplementary resources to effectively deliver the course to this large cohort.