Module MATH44220: Deep Learning, AI, and Data Analytics

Department: Mathematical Sciences

MATH44220: Deep Learning, AI, and Data Analytics

Prerequisites

• Statistical Inference, Machine Learning and Neural Networks

Corequisites

• None

Excluded Combination of Modules

• None

Aims

• To provide advanced methodological skills for the analysis of high-dimensional data, spanning the arc from the theoretical underpinning of the methods to the practical application on real data.
• To provide advanced methodological and practical knowledge in the field of deep learning and artificial intelligence, covering a wide range of the modelling and computational techniques ubiquitous in recent scientific and technological applications.

Content

• Term 1: Multivariate statistics; principal component analysis, cluster analysis, and related methods; Dimension reduction; regularised regression techniques.
• Term 2: Multilayer perceptrons; deep networks: CNNs, space, and computer vision; RNNs, time, and language processing; SGD and variants, dropout, etc.; network design, programming deep networks.
• Extensions: autoencoders, GANs, or reinforcement learning.

Learning Outcomes

• By the end of the module students will:
• Have gained a thorough understanding and working knowledge of advanced statistical methods for multivariate data;
• Have gained an appreciation of the wider concepts that such methods are built on, allowing straightforward understanding of yet unseen methods;
• Have gained intuition to distinguish supervised and unsupervised learning scenarios, and which specific methods to apply in particular situations;
• Have a systematic and coherent understanding of the mathematical theory underlying deep neural networks and their training;
• Have an understanding of the relationship of this theory to other statistical techniques;
• Be able to make appropriate modelling and algorithmic choices for a given problem or application;
• Be able to implement those choices using currently available software packages, and test their validity and performance;
• Have sufficient understanding and expertise to be able to expand their knowledge of theory and practice to encompass newly developed techniques and software.

• Students will have advanced mathematical skills in the following areas: modelling, optimisation, computation.

• Students will have advanced skills in the following areas: problem formulation and solution, synthesis of data, critical and analytical thinking, report writing, computer skills.

Modes of Teaching, Learning and Assessment and how these contribute to the learning outcomes of the module

• Lectures demonstrate what is required to be learned and the application of the theory to practical examples.
• Computer practicals consolidate the studied material and enhance practical understanding.
• Assignments for self-study develop problem-solving skills and enable students to test and develop their knowledge and understanding.
• Formative assessments provide feedback to guide students in the correct development of their knowledge and skills in preparation for the summative assessment.
• Written project reports assess the ability to implement the concepts introduced in the module using statistical software, to apply them in the analysis of a realistic problem, and to report scientific outputs in a clear and structured way.
• The end-of-year examinations assess the knowledge acquired and the ability to solve predictable and unpredictable problems.

Teaching Methods and Learning Hours

Summative Assessment

Formative Assessment:

Six written or electronic assignments to be assessed and returned. Other assignments are set for self-study and complete solutions are made available to students.

■ Attendance at all activities marked with this symbol will be monitored. Students who fail to attend these activities, or to complete the summative or formative assessment specified above, will be subject to the procedures defined in the University's General Regulation V, and may be required to leave the University