Aims and Fit of Module

This module aims to introduce students to a range of topics in the field of artificial neural networks while providing hands-on experience with some established works. It also seeks to highlight contemporary issues in neural computation, particularly in biologically-motivated computing, with an emphasis on multidisciplinary research. Additionally, the module underscores the importance of staying up-to-date in rapidly evolving areas of science and technology, equipping students with the necessary skills to do so. Finally, it prepares students to make well-reasoned decisions about the engineering of machine learning systems.

Learning outcomes

A. Describe the biological and historical developments of neural computation, and explain the nature and operation of Perceptron, MLP, CNN, Competitive learning, Oja learning, and SOM networks, including their applications and limitations. 
B. Apply foundational and advanced concepts in neural computation—including Perceptron, Multilayer Perceptron (MLP), Convolutional Neural Networks (CNN), Competitive Learning, Oja Learning, and Self-Organizing Maps (SOM)—to analyze, model, and solve emerging research challenges in machine learning and neural networks, including evaluating the suitability of different architectures and learning rules for specific tasks (e.g., classification, feature extraction, unsupervised learning) and justifying methodological choices based on theoretical principles and empirical performance.
C. Understand and apply ANN models for specific applications such as regression, prediction, and classification, reinforced by problem-solving experiences. 
D. Understand contemporary topics in artificial neural networks and deep neural networks, including CNN, and their advantages and applications. 
E. Apply modern machine learning concepts to analyze, design, and evaluate solutions for real-world problems. 
F. Utilize Matlab and relevant toolboxes for practical applications of ANNs.

Method of teaching and learning

The module is structured around a didactic component, which forms the core of the teaching through lecture-based instruction, supplemented by Q/A sessions and feedback. These lectures are further supported by tutorials and lab practicals to reinforce learning. Alongside this, the self-learning component encourages students to explore the subject materials independently, deepening their understanding through additional reading.  To assess comprehension, review exercises are integrated into the module, including two continuous assessments. These are preceded by supervised discussions and Q/A sessions during seminars to ensure clarity. Additionally, case studies are provided to help students contextualize the course material, bridging theory with real-world applications.