Machine Learning (2025/2026)
Published:
Machine learning is a field of artificial intelligence that focuses on developing algorithms and models that allow computers to learn from data and improve their performance without being explicitly programmed. It combines concepts from statistics, mathematics, and computer science to uncover patterns, make predictions, and support decision-making across diverse domains such as healthcare, finance, natural language processing, and robotics.
This course introduces students to the fundamental methods and principles of machine learning, progressing from basic concepts to advanced techniques:
LU1: Basics – Foundations of supervised learning, linear models, and regularisation strategies, including hands-on assignments to practice core ideas.
LU2: Classic Non-Linear Methods – Exploration of support vector machines, decision trees, and neural networks.
LU3: Ensembles – Introduction to bagging, boosting, and model distillation to enhance predictive performance.
LU4: Generalisation – Techniques for hyperparameter optimisation, evaluation metrics, and strategies to improve model robustness.
LU5: Unsupervised Learning – Dimensionality reduction (PCA, t-SNE) and clustering methods like k-means, paired with literature engagement.
LU6: Advanced Topics – Reinforcement learning, drawing on the work of Peter Abbeel and David Silver, culminating in a collaborative group project.
The course combines theoretical lectures with practical sessions led by teaching assistants. Students will engage in a mix of individual graded assignments, group projects, and a final oral exam. Through this structure, participants will gain both a solid theoretical understanding and practical experience in applying machine learning methods to real-world problems.