Skip to main content

Neural Networks and Deep Learning

Overview

  • Credit value: 15 credits at Level 7
  • Module convenor and tutor: Professor George Magoulas
  • Prerequisite: knowledge of mathematical concepts such as calculus and linear algebra; for MSc students, Applied Machine Learning
  • Assessment: a two-hour written examination (80%) and mid-term take-home assessment (20%)

Module description

Using a combination of lectures and lab work, this module covers neural networks and deep learning to give you knowledge of advanced features of various models and algorithms covering both theory and practice.

The module uses mathematical concepts, such as vector, matrices and their operations, functions and graphs, gradient and derivative, as well as trigonometry concepts, statistical concepts and the notion of probability, data structures, first-order and second-order optimisation methods and general algorithmic concepts.

INDICATIVE MODULE SYLLABUS

  • Shallow and deep neural network architectures and hybrid schemes
  • Supervised and unsupervised learning
  • Reinforcement learning
  • Optimisation methods for training and architecture search
  • Generalisation in neural networks and deep learning
  • Advanced topics and applications

Learning objectives

By the end of this module, you will be able to:

  • describe fundamental neural network architectures and learning algorithms
  • explain issues relating to the design and implementation of systems that employ neural networks and deep learning, and potential limitations
  • apply theoretical understanding of neural network architectures and learning algorithms to solve data modelling, classification and decision-making problems, justify the approach adopted and critically evaluate its effectiveness
  • recognise social, ethical and professional issues and risk involved in the application and use of neural networks and deep learning
  • demonstrate advanced knowledge of intelligent systems that combine neural networks and deep learning with other AI and machine learning paradigms and of the processes involved in their deployment
  • make a critical analysis of problem requirements relating to the application of neural networks and hybrid neural systems and of the necessary methods for training and evaluating them
  • demonstrate a comprehensive understanding of the principles and practices of designing intelligent systems that use neural networks and deep learning.