The first part of the course is on Feature Extraction and Machine Learning. The second part of the course is on basic neuroscience and how bioinspired algorithms from the brain have given rise to the field of Artificial Neural Networks (ANNs).
The Feature Extraction: combines classical theory on signal processing with modern machine learning (ML) Libraries for systematic feature engineering and analytics. Applications covered by this module are predictive maintenance, activity recognition, and spectrometric analysis. A student who successfully completes this part of the course will: 1) Be able to Develop Algorithms for time-series classification and time-series exogeneous regression. 2) Be able to Apply Systematic time-series feature engineering. 3) Be able to interpret time-series features 4) Be able to design and model in Python using libraries like sklearn and tsfresh.
Artificial Neural Networks: Knowledge of basic neuroscience - namely, component parts of the brain, components parts of the neuron, typical brain patterns observed in the EEG and Epilepsy. Analysis application and synthesis of the 4 common ANN models will be taught in the course. These are: The Artificial Neuron model, The Multi-Layer Perceptron model, The Self-Organising Map and the Radial Basis Function Network. A student who successfully completes this part of the course will: 1) Be able to perform hand calculations and develop Algorithms for The Artificial Neuron model, The Multi-Layer Perceptron model, The Self-Organising Map and the Radial Basis Function Network in Matlab programming language and interpret the algorithm performance. 2) Be able to identify, understand and interpret the functionality of: the component parts of the brain, components parts of the neuron, typical brain patterns observed in the EEG and Epilepsy.