Module overview
Machine Learning advances are revolutionising our world. At a fundamental level, Machine Learning deals with the extraction of useful information from large and complex datasets. There are now many applications, from the automatic understanding and processing of written text, the automatic detection of obstacles in stereo camera images in self driving cars or the recognition of human speech as in virtual digital assistants.
Machine learning is a broad discipline, requiring a basic understanding of many areas of science, from mathematical and statistical concepts to computational techniques. This module focusses on the fundamental concepts of modern Machine Learning for students studying engineering disciplines other than computer science and electronic engineering. The module will cover a range of fundamental principles, introduce state of the art techniques and will allow you to develop the practical skills to implement these techniques to analyses real world data to solve realistic engineering challenges.
Aims and Objectives
Learning Outcomes
Knowledge and Understanding
Having successfully completed this module, you will be able to demonstrate knowledge and understanding of:
- Formulate, derive and analyse machine learning methods and technologies using the appropriate mathematical foundations.
- Apply the underlying mathematical principles from probability, linear algebra and optimisation to solve data analysis problems.
Subject Specific Intellectual and Research Skills
Having successfully completed this module you will be able to:
- Design and implement solutions for complex machine learning problems using a scientific computing environment.
- Select and apply appropriate machine learning techniques to describe data in terms of explanatory models.
- Systematically apply machine learning to data in order to learn new patterns or concepts.
Full CEng Programme Level Learning Outcomes
Having successfully completed this module you will be able to:
- As part of the exam, students must demonstrate their ability to identify, analyse and evaluate sustainability issues that can arise from the use of data driven machine learning models.
- As part of the coursework, students must demonstrate their ability to select and apply appropriate machine learning techniques to model data in terms of explanatory models and to systematically apply machine learning to data in order to learn new patterns or concepts.
- As part of the coursework, students must demonstrate their ability to design and implement solutions for complex machine learning problems using a scientific computing environment.
- As part of the exam, students must demonstrate their ability to identify, analyse and evaluate security issues that can arise from the use of data driven machine learning models.
- As part of the exam, students must demonstrate their ability to identify equality, diversity and inclusion issues that can arise from the use of data driven machine learning models.
- As part of the exam, students must demonstrate their ability to apply the underlying mathematical principles from probability, linear algebra and optimisation to data analysis problems.
- As part of the exam, students must demonstrate their ability to identify, analyse and evaluate ethics issues that can arise from the use of data driven machine learning models.
- As part of the exam, students must demonstrate their ability to formulate, derive and analyse machine learning methods and technologies using the appropriate mathematical foundations.
- As part of the coursework and their exam, students must demonstrate their ability to appraise several state of the art machine learning methods, their implementation and limitations when applied to complex real world problems
Subject Specific Practical Skills
Having successfully completed this module you will be able to:
- Critically appraise several state-of-the-art machine learning methods, their implementation and limitations when applied to complex real-world problems.
Syllabus
Mathematical and statistical foundations
- Linear algebra and vector spaces
- Cost functions, optimisation and stochastic gradient descend
- Discrete and continuous random variables
- Distributions and conditional distributions
- Hypothesis testing, likelihood optimisation and Bayesian methods
Machine Learning basics:
- Supervised vs. unsupervised learning
- Classification, regression, clustering and dimensionality reduction
- Performance measures and empirical risk minimization
- Overfitting, training, testing, validating
- Bias variance trade-offs
- Classical machine learning methods and artificial neural networks
Modern deep learning models:
- Deep feedforward and recurrent models
- Auto-encoders
- Generative models
Application examples:
- Multivariate data
- Image data
- Time series data
Common scientific computing tools for machine learning:
- Data handling tools
- Machine learning tools
- Visualisation tools
Societal issues:
- Data biases in machine learning and their impact on equality, diversity and inclusion
- Machine learning ethics
- Sustainability of machine learning solutions
- Security of machine learning systems
Learning and Teaching
Teaching and learning methods
Lectures, tutorials, computing labs, data analysis and computing exercises, and guided self-study
Type | Hours |
---|---|
Assessment tasks | 40 |
Wider reading or practice | 38 |
Tutorial | 15 |
Lecture | 12 |
Specialist Laboratory | 9 |
Private study hours | 36 |
Total study time | 150 |
Assessment
Summative
This is how we’ll formally assess what you have learned in this module.
Method | Percentage contribution |
---|---|
Coursework | 50% |
Exam | 50% |
Referral
This is how we’ll assess you if you don’t meet the criteria to pass this module.
Method | Percentage contribution |
---|---|
Coursework | 50% |
Exam | 50% |
Repeat
An internal repeat is where you take all of your modules again, including any you passed. An external repeat is where you only re-take the modules you failed.
Method | Percentage contribution |
---|---|
Exam | 50% |
Coursework | 50% |