CS4S772 - Deep Learning 01 Sep 2021 - 31 Aug 2028 | Version 2

Associated Module Information

Module Code: CS4S772
Module Title: Deep Learning
Faculty: Faculty of Computing, Engineering and Science
Faculty Group: Computing and Mathematical Sciences
Faculty Sub Group: Computer Science
Module Leader: Mabrouka Abuhmida
Module Team: Andrew Ware
First Intended Intake: SEP 2019 Final Year of Intake:
Date Closed:
Credit Value: 20 Credit Level: 7
Language: English
Percentage of Module Taught in Welsh: 0
Equivalent Module:
HECOS codes: 101020 - computer games programming
HECOS Code Weighting: 100

Document Version Information

Version 2
Valid From 01 Sep 2021
Valid To 31 Aug 2028

Module Aims

To introduce deep learning, approaches to their design and development, areas of application and available tools.

Content Summary

Deep feedforward networks, convolutional neural networks, deep belief networks, recurrent neural networks, reinforcement learning, classification, prediction, regularisation, optimisation algorithms, sequence modelling, methodology, practical examples: natural language processing, speech recognition, computer vision, online recommendation systems, bioinformatics and videogames.

Learning and Teaching Methods

Activity Type Hours
Lecture 24
Practical classes and workshops 24
Independent Study 80
Directed Study 72
Total Hours Selected 200

Learning Outcomes

# Learning Outcome
LO1 To demonstrate knowledge and comprehension in the explanation of deep learning methods, tools and models.
LO2 To demonstrate discernment in the effective application of deep learning to design and implement a problem dependent solution utilising appropriate deep learning methods, tools and models.

Module Requisites

N/A

Assessment Criteria

Assessment Category Assessment Type Description Duration Word Count Weight (%) Best of? Pass Mark
Asynchronous Assessment Practical Written Work 1 Apply practical knowledge of deep learning techniques to solve a constrained problem and report on findings in the style of an academic paper. 0 2000 50 No 40
Asynchronous Assessment Practical Written Work 2 Apply practical knowledge of deep learning techniques to solve a constrained problem and report on findings in the style of an academic poster supported by an oral presentation. 0 2000 50 40

Assessment Matrix

Assessment Type Learning Outcomes
LO1 LO2
Practical Written Work 1 ✔ ✔
Practical Written Work 2 ✔ ✔

Reading List

Deep Learning with Python. Francois Chollet (2017), Manning Publications, ISBN-13: 978-1617294433

Practical Convolutional Neural Networks.Mohit Sewak, Md. Rezaul Karim, Pradeep Pujari (2018), Packt Publishing, ISBN-13: 978-1788392303

Python Deep Learning Projects. Matthew Lamons, Rahul Kumar, Abhishek Nagaraja (2018), Packt Publishing, ISBN-13: 978-1788997096

Deep Learning (Adaptive Computation and Machine Learning Series). Ian Goodfellow, Yoshua Bengio, Aaron Courville, Francis Bach (2017), MIT Press, ISBN-13: 978-0262035613

Introduction to Deep Learning. Sandro Skansi (2018) Springer, ISBN 978-3-319-73003-5