Machine Learning (Elective)

The aims of this module are to provide learners with an understanding of key concepts, techniques, and practical skills in the field of machine learning. With this grounding learners will be trained to push further and communicate results to a variety of audiences (technical, adept, expert). A focus on the interpretability, repeatability and reproducibility of work to instil trust in results presented will be at the forefront. Key issues in ethics will also be covered as these societal challenges provide varying levels of social acceptability in application areas.

Theoretical base for machine learning

• Probability and statistics
• Linear mathematics
• Supervised learning vs unsupervised learning

Considerations for data in machine learning

• Data types and structures
• Pre-processing and cleaning
• Feature extraction and engineering
• Data augmentation
• Data imputation

Unsupervised learning

• K-means clustering
• Agglomerative clustering
• Density-Based Spatial Clustering of Applications with Noise
• Dimensionality reduction (PCA and UMAPS)

Model fitting

• Cross validation
• Over fitting and under fitting
• Precision and accuracy
• F1 score

Supervised learning

• Linear regression
• Logistic regression
• Decision trees and random forests
• K-nearest neighbours
• Support vector machines

Ethics

• Case studies and report writing
• Transparency and interpretability of results
• Practical limitations due to ethical concerns