Artificial Intelligence (PgCert)

Program Overview

---

Artificial Intelligence (PgCert)

Overview

The PG Cert in Artificial Intelligence (AI) is aimed as a starting point to prepare students to embark on an industrial career or further research studies, with knowledge and skills in AI mathematics, knowledge representation and reasoning, machine learning, computer vision, natural language processing, and data analytics.

Course Structure

Taught modules will be running in block mode.

Modules

Core Modules

• Machine Learning (20 credits)
• Computer Vision (20 credits)
• Foundations of AI (20 credits)

Entrance Requirements

Graduate

Normally a 2.1 Honours degree or equivalent qualification acceptable to the University in Computer Science, Software Engineering, Electrical and/or Electronic Engineering, Mathematics with Computer Science, Physics with Computer Science or a related discipline. Applicants must normally have achieved 2:1 standard or above in relevant modules.

Applicants who hold a 2.2 Honours degree and a Master’s degree (or equivalent qualifications acceptable to the University) in one of the above disciplines will be considered on a case-by-case basis.

All applicants will be expected to have mathematical ability and significant programming experience as evidenced either through the content of their primary degree or through another appropriate formal qualification.

Applications may be considered from those who do not meet the above requirements but can provide evidence of recent relevant technical experience in industry, for example, in programming.

International Students

Our country/region pages include information on entry requirements, tuition fees, scholarships, student profiles, upcoming events and contacts for your country/region.

English Language Requirements

Evidence of an IELTS* score of 6.5, with not less than 5.5 in any component, or an equivalent qualification acceptable to the University is required. *taken within the last 2 years

Tuition Fees

Northern Ireland (NI) 1 | £2,434
---|---
Republic of Ireland (ROI) 2 | £2,434
England, Scotland or Wales (GB) 1 | £3,083
EU Other 3 | £8,600
International | £8,600

Additional Course Costs

Students may incur additional costs for small items of clothing and/or equipment necessary for lab or field work.

Career Prospects

The objective of this course is to produce highly trained and desirable graduates who are well-prepared for the rapidly-evolving field of AI technology.

Prizes and Awards

Teachers working on classroom-based dissertation projects may apply for the Northern Ireland Centre for Educational Research (NICER) award.

Graduate Plus/Future Ready Award for extra-curricular skills

In addition to your degree programme, at Queen's you can have the opportunity to gain wider life, academic and employability skills.

International Scholarships

Information on scholarships for international students, is available at

Machine Learning

Overview

Core concepts in machine learning via linear regression
• The machine learning workflow; design and analysis of machine learning experiments
• Linear regression: least-squares and maximum likelihood
• Generalisation: overfitting, regularisation and the bias-variance trade-off
Classification
• Classification algorithms: k-NN, logistic regression, decision trees, support vector machine,
ensemble learning
• Evaluation metrics for classification models
• Explainable AI (XAI): feature attribution methods for black-box algorithms
Bayesian machine learning and probabilistic programming
• Bayesian approach to machine learning; Bayesian linear regression
• Bayesian non-parametric models: Gaussian Process regression
• Probabilistic programming; Markov Chain Monte Carlo methods and diagnostics
Unsupervised learning
• Clustering algorithms: k-means, hierarchical clustering, density-based clustering
• Evaluation metrics for clustering algorithms
• Dimensionality reduction: PCA and PLS
Case studies

Reading List:
Stuart Russell and Peter Norvig. Artificial Intelligence: A Modern Approach, 2020, Pearson
Tom M. Mitchell. Machine Learning, McGraw-Hill

Learning Outcomes

Successful students will be able to:
1. Demonstrate critical understanding of the theory underpinning core concepts and algorithms in
machine learning
2. Evaluate and compare supervised and unsupervised learning algorithms on problems involving
real datasets
3. Diagnose and rectify common problems that affect the performance of machine learning
algorithms
4. Design machine learning experiments and justify the procedures employed

Skills

Ability to identify problems that can be solved using machine learning methods. Ability to apply suitable classical machine learning algorithms and software packages to solve real-world problems. Ability to evaluate and compare the performance of machine learning methods for a given problem.
Present and discuss the results of machine learning methods and propose appropriate improvements to methods.

Coursework

100%

Examination

0%

Practical

0%

Credits

20

Module Code

ECS8051

Teaching Period

Autumn

Duration

4 weeks

Pre-requisite

No

Core/Optional

Core

Computer Vision

Overview

Traditional Computer Vision
(a selection of the following)
Introduction
Image acquisition; Image representations; Image resolution, sampling and quantisation; Colour models

Representation for Matching and Recognition
Histograms, thresholding, enhancement; Convolution and filtering
Scale Invariant Feature Transform (SIFT)
Hough transforms
Geometric hashing
Image representation and filtering in the frequency domain; JPEG and MPEG compression

Neural networks
Loss functions and stochastic gradient descent;
Backpropagation; Architecture of Neural Network and different activation functions;
Issues with training Neural Networks
Autograd; Hyperparameter optimisation

Deep Learning for Computer Vision
(a selection of the following)
Convolutional Neural Networks: image classification
Generative adversarial networks: image generation
Residual Networks (ResNet)
YOLO: object detection
Vision Transformer

Case studies

Reading List:
Szeliski, R. Computer Vision: Algorithms and Applications. 2011, Springer.
Ian Goodfellow, Yoshua Bengio and Aaron Courville. Deep Learning, 2017, MIT Press.

Learning Outcomes

Successful students will be able to:
1. Utilise principles and theories of Computer Vision in selected scenarios,
2. Design and implement Computer Vision techniques to solve practical problems using software tools and libraries.
3. Evaluate potential solutions to problems using image and video data.

Skills

Ability to utilise deep learning algorithms and techniques to solve real-world computer vision challenges. Creativity in obtaining image/video data from recognised repositories. Ability to utilise existing libraries and packages for implementing appropriate machine learning models for a given computer vision task. Recognising patterns in image/video data in a convincing way.

Coursework

100%

Examination

0%

Practical

0%

Credits

20

Module Code

ECS8053

Teaching Period

Spring

Duration

4 weeks

Pre-requisite

No

Core/Optional

Core

Foundations of AI

Overview

Constraint satisfaction

Markov decision processes

Probability and Statistics
Random variables
Conditional and joint distributions
Variance and expectation
Bayes Theorem and its applications
Law of large numbers and the Multivariate Gaussian distribution

Calculus:
Differential and integral calculus, partial derivatives, vector-values functions, directional gradient

Optimisation
Convexity
1-D minimisation
Gradient methods in higher dimensions

Linear Algebra
Using matrices to find solutions of linear equations
Properties of matrices and vector spaces
Eigenvalues, eigenvectors and singular value decomposition

Learning Outcomes

Successful students will be able to:
1. Demonstrate knowledge and critical understanding of topics from linear algebra, calculus, probability, statistics and optimisation that are required to apply in AI
2. Relate the mathematics topics to

Skills

Ability to utilise fundamental mathematics underlying AI and identify the most suitable modelling, optimisation, factorisation, and transformation approach for a given problem.

Coursework

100%

Examination

0%

Practical

0%

Credits

20

Module Code

ECS8050

Teaching Period

Autumn

Duration

4 weeks

Pre-requisite

No

Core/Optional

Core

Program Outline

---

Degree Overview:

---

The Artificial Intelligence (PgCert)

• Aims to provide students with the knowledge and skills necessary to embark on an industrial career or further research studies in AI.
• Focuses on mathematics, knowledge representation and reasoning, machine learning, computer vision, natural language processing, and data analytics.
• Students will gain experience in applying AI knowledge and skills to develop AI systems and applications.
• Provides an introduction to core topics, enabling students to understand the range of topics and acquire skills associated with AI systems and applications.

---

Target Audience:

• Analytical, curious, technical, and ambitious individuals
• Those who want to expand their horizons in the field of AI
• Those who are aware of the growing demand for AI skills and want to advance their careers in this area
• Ideally, Computing graduates with strong programming skills and a solid background in mathematics.

---

Outline:

---

Course Structure:

• Taught modules run in block mode.
• Semester length: 1 year
• Places available: To be confirmed (Part Time)

---

Modules:

• Computer Vision
• Deep neural networks (DNNs) and modern computer vision approaches
• DNN models for various computer vision tasks
• Current topics of computer vision
• Develop ability to use DNN models to solve real-world computer vision challenges
• Ability to obtain image/video data from recognized repositories
• Ability to utilize existing libraries and packages for implementing appropriate DNN models for a given computer vision task
• Foundations of AI
• Fundamental mathematics underlying AI including probability and statistics, calculus, algebra, and optimization
• Develop a sound understanding of the fundamentals
• Develop ability to utilize them to understand and explain various AI techniques
• Identify the most suitable modeling, optimization, factorisation, and transformation approach for a given problem
• Machine Learning
• Different types of machine learning and various algorithms of each type
• Provide a systematic understanding of machine learning as a subject area
• Develop your ability to:
• Identify problems that can be solved using machine learning methods
• Apply suitable machine learning algorithms and software packages to solve real-world problems
• Evaluate and compare the performance of machine learning methods for a given problem
• Present and discuss the results of machine learning methods and propose appropriate improvements to methods

---

Assessment:

• Assessments associated with the course are outlined below:
• Awarding of the qualifications is based on continuous assessment of coursework.
• Assessment of modules is based solely on submitted work related to private, individual study.

---

Teaching:

• Learning Opportunities:
• Academic Team: Specialists in each subject area who bring a wealth of up-to-date knowledge to the course.
• Extensive research experience combined with projects offers you the perfect environment to study AI.
• English Language Support: The school offers support on the use of English in academic writing, to help you in your future career.
• Modules: Composed of three distinct modules, each intended to progressively enhance your knowledge, comprehension, and proficiencies in AI.
• Project-based assessment conducted following the taught period.
• To supplement the onsite sessions, there are substantial digital resources, such as pre-lecture videos and video-ed labs, with online support sessions available beyond the onsite teaching periods.
• Learning and Teaching Methods:
• Lectures
• Labs
• Group work
• Project work
• Online resources

---

Careers:

• Employers:
• BT, BBC, PwC, Kainos, Datactics,
• Microsoft, Google, Facebook, Oosto (formerly Anyvision), etc
• Career Development:
• A thought leader in AI, showcasing technological advancements through research.
• Working for some of the largest companies on the planet.
• Advising government policy.

---

Other:

• Industry Links:
Developed in direct response to industry need, this course will provide the building blocks required for you to step into a career in AI.
• World Class Facilities:
Most lectures and lab-based activities are in the Computer Science Building, a state-of-the-art facility with large well-equipped computing labs, including a dedicated AI Lab, and formal and informal student spaces which support a high level of group and project work.
• Internationally Renowned Experts:
The teaching team are specialists in each subject area and bring a wealth of up-to-date knowledge to the course. They have extensive research experience in their subject area and are noted for their research output.
• Student Experience:
The programme development team has experience in AI programme design at MSc level. This programme is newly designed to minimize module overlap, maximize employment relevancy and content recency; to consider knowledge/skill longevity and between-year continuity; to be free of legacy issues (existing course provision, staff). Four new AI staff members are recruited to best match the new design.

---

Tuition Fees and Payment Information:

---

Northern Ireland (NI)

• Free for DfE Funded students (see below): This indicates that the tuition fees for students residing in Northern Ireland who are funded by the Department for the Economy (DfE) are waived.

---

Republic of Ireland (ROI)

• N/A: This suggests that there are no specific tuition fees mentioned for students from the Republic of Ireland.

---

England, Scotland, or Wales (GB)

• N/A: Similar to ROI students, there is no mention of specific tuition fees for those residing in England, Scotland, or Wales.

---

EU Other

• N/A: No information is provided regarding tuition fees for students from other European Union countries besides ROI.

---

International

• N/A: The text doesn't specify tuition fees for international students.

---

Additional Course Costs

• **Students may incur additional costs for small items of clothing and/or equipment necessary for lab or field work.

---

All Students

• **Depending on the programme of study, there may be extra costs which are not covered by tuition fees, which students will need to consider when planning their studies.

• **The text further elaborates on potential costs such as recommended textbooks, printing, photocopying, memory sticks, and graduation ceremonies.

---

International Scholarships

• **Information on scholarships for international students, is available at www.qub.ac.uk/Study/international-students/international-scholarships.

---

Please Note:

• While the text mentions "DfE Funded students" and "EU Other" students, it doesn't elaborate on the specific criteria or requirements for these categories.
• The information provided does not include the actual cost of tuition fees in any specific currency.