There are no pre-requisites for this module to be added

This module aims to extend the Fundaments of machine learning module. In particular, students will gain a deeper knowledge of modern machine learning algorithms, with an emphasis of neural networks forimage analysis tasks (e.g., classification, regression, segmentation). The syllabus of this moduleincludes:- Probability theory: introduction, discrete and continuous probability function, Bayesian classifier,information theory.- Image analysis: colour spaces,convolution, edge detection, image representation, image descriptors- Neural networks: multi-layer perceptron, convolutional layers, initialisation, optimisersIn the last part of the module, the students will learn about state-of-the-art deep neural networkarchitectures, putting them in the context of classification and regression tasks.

Upon completion of this module you will be able to

LO1: LO1: Critically evaluate the use of deep learning algorithms

LO2: LO2: Develop new skills in image analysis and natural language processing using deep networks

LO3: LO3: Critically evaluate the different layers in a deep neural network

LO4: LO4: Critically evaluate the different loss functions and optimisers

LO5: LO5: Design, develop, and evaluate customised deep neural networks applications in Python and PyTorch

LO6: LO6: Critically report the results in a scientific form and draw appropriate conclusions.