Preface

This is a module of two halves. The first half contains a brief review of some basic ideas in statistics that will lead us on to key concepts in machine learning (ML). We will be focussing on various common algorithms, when they are useful and how to implement them in R. The second half, in Epiphany term, will be a more detailed look at mechanics of optimisation and its importance to machine learning and moving onto neural networks and other more advanced techniques.

By the end of this term, I want you to be able to:

  1. Understand some common terms in machine learning.

  2. Be able to use R to utilise some of the most common machine learning techniques.

  3. Be able to evaluate the performance of a machine learning model.

  4. Realise that there are many difficult questions that arise in practice and these are lurking behind the scenes in their R implementations.

I have chosen to deliver the material in four ways: this complete set of lecture notes, handwritten notes in lectures to highlight key points, lecture slides to give more realistic examples and the four practical sessions where you will be able to implement the techniques in R. This term will be focussed on machine learning techniques in general and will not be specifically looking at neural networks.

The general structure for this half of the course will be as follows:

  1. key concepts in statistics for ML,
  2. fundamental ML concepts,
  3. classification techniques,
  4. regression techniques,
  5. making use of multiple models,
  6. making ML models more interpretable,
  7. unsupervised learning.

This half of the course will be supplemented by four formative assignments and four practical sessions. For the R elements, you can use your own installations of R or utilise the GitHub Codespace version that I have set up especially for this course. The latter will allow you to run R code in the cloud without having to install anything on your own machine. Here’s a link to the Codespace set-up page:

https://github.com/codespaces/new/johnpaulgosling/mlnn_docker?quickstart=1

Formal, summative assessment has two elements: there will be a practical exam in January and a written exam at the end of the academic year. The practical exam will be a test of your ability to implement the techniques in R and to evaluate the performance of a model. The written exam will be a test of your understanding of the theory behind the techniques.