Machine Learning

This module provides a broad view of machine learning and statistical pattern recognition.

You will learn several techniques, including supervised learning (e.g. generative and discriminative learning, parametric and non-parametric learning), unsupervised learning (e.g. clustering), and theoretical aspects of machine learning (e.g. bias, variance).

The module will also discuss recent applications of machine learning to areas of interest to data scientists.

Upon successful completion of this module, you will be able to:

  1. Analyse and explain fundamental machine learning concepts and a number of machine learning algorithms.
  2. Analyse the difference between different machine learning problems (e.g. supervised/unsupervised, classification/regression, clustering/dimensionality reduction).
  3. Select appropriate feature representations for different types of data.
  4. Select and implement appropriate machine learning algorithms for a particular dataset and problem.
  5. Apply and evaluate standard machine learning methods on data.

Topics covered

  • Introduction to Machine Learning
  • Classification
  • Regression
  • Model Improvement
  • Unsupervised Learning
  • Ensemble Methods
  • Neural Networks and Deep Learning
  • Working with Timeseries
  • Probabilistic Modelling
  • Ethics and Sustainability

Credits

15 (150 hours)
Summative coursework (30%)

Assessment

  • Written examination (70%)