This course is the next step beyond our introductory machine learning course and teaches students about modern techniques for machine learning with high-dimensional image and sequence (time-series) data, and the underlying computational structures for such systems.

Three hours per week.

None

None

Examination 80%, Written Assignment 15%, Set Exercise 5%.

Reassessments are normally available for all courses, except those which contribute to the Honours classification. For non-Honours courses, students are offered reassessment in all or any of the components of assessment if the satisfactory (threshold) grade for the overall course is not achieved at the first attempt. This is normally grade D3 for undergraduate students and grade C3 for postgraduate students. Exceptionally it may not be possible to offer reassessment of some coursework items, in which case the mark achieved at the first attempt will be counted towards the final course grade. Any such exceptions for this course are described below. 

The coursework cannot be redone because the feedback provided to the students after the original coursework would give any students redoing the coursework an unfair advantage.

The aim of this course is to go beyond our introductory machine learning course, and teach students about modern techniques for machine learning with high-dimensional image and sequence (time-series) data, and the underlying computational structures for such systems. Teach the students about managing large data sets, and the engineering pipelines for large-scale machine learning tasks.  In this course, students will learn the foundations of deep learning and dynamic models for time-series analysis.

By the end of this course students will be able to:

1. Understand the major technology trends in advanced machine learning;

2. Build, train and apply fully connected deep neural networks;

3. Know how to implement efficient, vectorised neural networks in python and understand the underlying backends;

4. Apply deep learning methods to new applications;

5. Understand the machine learning pipeline, and engineering aspects of training data collation, and the importance of unlabelled data.

Students must submit at least 75% by weight of the components (including examinations) of the course's summative assessment.