Machine learning ST3189

This course covers a wider range of such model based and algorithmic machine learning methods, illustrated in various real-world applications and datasets. At the same time, the theoretical foundation of the methodology is presented is some cases.

Prerequisites

If taken as part of a BSc degree, the following courses must be passed before this course may be attempted:

  • ST104a Statistics 1
  • ST104b Statistics 2
  • MT105a Mathematics 1 with MT105b Mathematics 2 or MT1174 Calculus. 

Topics covered 

  • Linear regression and regularisation (via least squares and maximum likelihood)
  • Bayesian Inference
  • Classification
  • Resampling methods
  • Clustering
  • Non-linear models
  • Tree-based methods
  • Support Vector Machines
  • Random forests
  • Gaussian Processes

Learning outcomes

At the end of the course and having completed the essential reading and activities students should be able to:

  • develop an understanding of the process to learn from data
  • be familiar with a wide variety of algorithmic and model based methods to extract information from data 
  • apply and evaluate suitable methods to various datasets by model selection and predictive performance evaluation.  

Assessment

An individual case study piece of coursework (30%) and a 2 hr unseen written examination (70%). 

The coursework will involve several computer exercises in R (no prior knowledge is required). 

Essential reading

Rogers S. and Girolami M. A First Course in Machine Learning, Chapman & Hall/CRC Press, second edition (2011) [ISBN 9781498738484].

James G., Witten D., Hastie T. and Tibshirani R. An introduction to Statistical Learning: with Applications in R, Springer (2013) [ISBN 9781461471387]