WM9B7-15 Artificial Intelligence & Deep Learning
Introductory description
In today's world, artificial intelligence and data science are powering innovation in virtually all industries and domains. The ability to build machines, and algorithms, that are able to reason and make decisions autonomously offers not only huge benefits to modern business, but to society as a whole. This module provides a hands-on exposure to the practice of developing AI/machine learning algorithms and implementing them in a variety of problem sets and datasets
Module aims
This module aims to enable participants to select, implement and evaluate deep learning algorithms in data science and artificial intelligence. In particular, the module highlights several of the most common, and in-demand, modern algorithms including recurrent, convolutional and other neural networks. Alongside technical knowledge, participants should develop an understanding of the applicability of different types of artificial intelligence & machine learning to common problems, and best practice for data science and artificial intelligence projects.
Outline syllabus
This is an indicative module outline only to give an indication of the sort of topics that may be covered. Actual sessions held may differ.
Core concepts of Artificial Intelligence & Deep Learning; Data pre-processing & engineering; Optimisation algorithms (SGD, Adam, etc.); Artificial Neural Networks (ANN); autoencoders; Convolutional Neural Networks (CNN); Recurrent Neural Networks (RNN) & Long-Short Term Memory (LSTM); transformer models; Q-learning; Bayesian Neural Networks (BNN); Variational Autoencoders (VAE); Generative Adversarial Networks (GAN); transfer learning; Siamese networks; self-supervised learning; Model training and evaluation.
Learning outcomes
By the end of the module, students should be able to:
- Interpret and evaluate various use-cases and the applicability of artificial intelligence and deep learning.
- Adopt best practices for data processing and engineering for artificial intelligence and deep learning models.
- Implement, interpret and critique current, professional standard learning models.
- Automate deployment-ready deep learning pipelines and algorithms.
- Evaluate and interpret the results of deep learning models and tune them to optimise performance.
Interdisciplinary
In particular, combining computer science and mathematics/statistics
International
International demand remains high for graduates with the skills incorporated in this module
Subject specific skills
Artifical intelligence, deep learning, statistics, machine learning, software development, data analysis
Transferable skills
Programming, statistics and modelling, team work, critical analysis
Study time
| Type | Required |
|---|---|
| Lectures | 12 sessions of 1 hour (8%) |
| Seminars | 6 sessions of 1 hour (4%) |
| Supervised practical classes | 12 sessions of 1 hour (8%) |
| Online learning (independent) | 60 sessions of 1 hour (40%) |
| Assessment | 60 hours (40%) |
| Total | 150 hours |
Private study description
No private study requirements defined for this module.
Costs
No further costs have been identified for this module.
You do not need to pass all assessment components to pass the module.
Assessment group A2
| Weighting | Study time | Eligible for self-certification | |
|---|---|---|---|
| Algorithm Development | 20% | 10 hours | No |
|
Students work on a real dataset to apply learning and present their results |
|||
| Assignment | 80% | 50 hours | Yes (extension) |
|
Essay on artificial intelligence and deep learning topics (including original code creation) |
|||
Assessment group R2
| Weighting | Study time | Eligible for self-certification | |
|---|---|---|---|
| Assignment | 100% | Yes (extension) | |
|
PMA on artificial intelligence and deep learning topics (including original code creation) |
|||
Feedback on assessment
Verbal feedback for in-module components; written feedback and annotated scripts for post-module work
Pre-requisites
Pre-requisite to be added once that module has been approved (Data Engineering with Python)
There is currently no information about the courses for which this module is core or optional.