Last modified: 31 May 2022 13:05
In this course we will discuss modern methods of machine learning, such as decision trees, regression, Markov models, Bayesian approaches, Nearest Neighbours, random forests, support vector machines and neural networks.
Great emphasis will be given to the actual application of all these methods to small and large data sets.
| Study Type | Postgraduate | Level | 5 |
|---|---|---|---|
| Term | First Term | Credit Points | 15 credits (7.5 ECTS credits) |
| Campus | Aberdeen | Sustained Study | No |
| Co-ordinators |
|
||
In this course we will discuss modern methods of machine learning, such as decision trees, regression, Markov models, Bayesian approaches, Nearest Neighbours, random forests, support vector machines and neural networks.
The course is very practical and great emphasis will be on the actual application of all these methods to small and large data sets.
First, we will use high level functions that perform these analyses in an automated way and we will focus on data preparation and interpretation of the results.
As the course progresses, we will move to more advanced techniques and construct for example neural networks from scratch. We will learn how to perform network surgery to benefit from pertained networks and achieve maximal accuracy and efficiency in our training and for the predictions.
Information on contact teaching time is available from the course guide.
1st Attempt
Practical 50%
MCQ Online Test 50%
Alternative Resit Arrangements
Resubmission of failed elements (pass mark carried forward)
There are no assessments for this course.
| Knowledge Level | Thinking Skill | Outcome |
|---|---|---|
| Procedural | Apply | Knowledge of modelling techniques |
| Reflection | Understand | Understanding difference between stochastic and deterministic processes |
| Reflection | Apply | Understanding computation processes |
We have detected that you are have compatibility mode enabled or are using an old version of Internet Explorer. You either need to switch off compatibility mode for this site or upgrade your browser.
