Module Code - Title:
CS4287
-
NEURAL COMPUTING
Year Last Offered:
2025/6
Hours Per Week:
Grading Type:
N
Prerequisite Modules:
Rationale and Purpose of the Module:
The objective of this module is to equip students with the fundamental knowledge and techniques necessary to effectively apply artificial neural networks to a wide variety of problems in a machine learning context, and correctly interpret the results.
Syllabus:
Introduction to the computational model of a neuron.
Models of Learning: Hebian, Boltzman, supervised, unsupervised, and reinforcement learning.
Learning in the Perceptron and its limitations.
Backpropagation in the Multilayer Perceptron.
Cross validation, generalisation, over-fitting, and analysis of the output.
Hopfield networks.
Deep learning paradigms such as Convolutional Neural Networks, Long Short Term Memory, and Recurrent Neural Networks. Concepts such as Dropout and Batch Normalisation will be introduced.
Reinforcement Learning paradigms such as Temporal Difference Methods and their implementation on neural architectures.
Applications of neural computing to a wide range of domains. Examples include object identification and recognition in computer vision, financial prediction, synthesis of texts, simulated robot control. Implementations will generally use third party APIs.
Topics: Radial-Basis Function networks, Self-Organising maps, Support Vector Machines.
Learning Outcomes:
Cognitive (Knowledge, Understanding, Application, Analysis, Evaluation, Synthesis)
On successful completion of this module, students should be able to:
1. Describe the architecture and learning methods of shallow and deep networks.
2. Design, build, train, and evaluate both shallow and deep networks.
3. Select an appropriate neural paradigm for the problem at hand.
4. Compare and contrast neural computing with other approaches in the field of machine learning.
Affective (Attitudes and Values)
On successful completion of this module, students should be able to:
1. Demonstrate an appreciation of the complexity of neural computing.
2. Defend the choice of neural computing method in a particular context.
Psychomotor (Physical Skills)
N/A
How the Module will be Taught and what will be the Learning Experiences of the Students:
Students will be proactive in lab-based project settings with a focus on collaborative
Research Findings Incorporated in to the Syllabus (If Relevant):
Prime Texts:
Simon Haykin (2009)
Neural Networks and Machine Learning (3rd Edition)
, Pearson.
Ian Goodfellow, Yoshua Bengio, and Aaron Courville (2016)
Deep Learning
, The MIT Press
Other Relevant Texts:
Aurélien Géron (2017)
Hands-On Machine Learning with Scikit-Learn and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems
, O'Reilly
Li Deng and Dong Yu (2014)
Deep Learning: Methods and Applications.
, Now Publishers Inc
Programme(s) in which this Module is Offered:
BSCOSYUFA - COMPUTER SYSTEMS
BSCGDEUFA - COMPUTER GAMES DEVELOPMENT
Semester(s) Module is Offered:
Autumn
Module Leader:
j.j.collins@ul.ie