| Unit name | Applied Deep Learning (Teaching Unit) |
|---|---|
| Unit code | COMSM0045 |
| Credit points | 0 |
| Level of study | M/7 |
| Teaching block(s) |
Teaching Block 1 (weeks 1 - 12) |
| Unit director | Dr. Wray |
| Open unit status | Not open |
| Units you must take before you take this one (pre-requisite units) |
COMSM30035 Machine Learning (Teaching Unit) or equivalent. |
| Units you must take alongside this one (co-requisite units) |
EITHER COMSM0158 Advanced Topics in Computer Science (Examination assessment, 20 credits) OR COMSM0155 Applied Deep Learning (20 credits). Please note: COMSM0043 is the Teaching Unit for the Applied Deep Learning option. Students taking this unit choose to be assessed by EITHER the MAJOR 20 credit unit (COMSM0155) OR as part of the Advanced Topics in Computer Science MINOR 20 credit examination unit. Students select the form of assessment to be taken by enrolling on the appropriate co-requisite assessment unit. |
| Units you may not take alongside this one |
None |
| School/department | School of Computer Science |
| Faculty | Faculty of Engineering |
Why is this unit important?
Deep learning is the current state-of-the-art methodology for Machine Learning and Artificial Intelligence. This unit introduces the students to the latest deep architectures, as well as learning approaches to optimising these for a variety of classification and regression problems with an applied focus.
The unit will first focus on a theoretical foundation of Deep Learning before introducing different models and learning methods with an applied focus – students will explore the area in a hands on approach in labs applying various tools and techniques on research data.
How does this unit fit into your programme of study
This is an optional unit that can be taken during TB1 in Year 4. This allows for students to build on knowledge learnt within the first 3 years of their study to learn about the state-of-the-art in Deep Learning.
An overview of content
The unit paves the path from understanding the fundamentals of previously and currently used deep learning methods (such as multi-layer perceptrons and deeper neural networks) through to training and optimisation as well as evaluation of learnt outcomes. The unit’s approach is hands-on, focusing on the ‘howto’ while covering the basic theoretical foundations.
How will students, personally, be different as a result of the unit
Students will be aware of state of the art of Deep Learning and understand the benefits & limitations to these methods, and how these methods can be applied to a variety of machine learning problems.
Learning Outcomes
On successful completion of this unit, ALL students (both MAJOR and MINOR) will be able to:
1. Identify the opportunities and challenges that deep architectures bring to machine learning tasks such as classification and regression.
2. Analyse and discuss the role of a variety of optimization approaches in parameter training for deep architectures.
3. Setup, train and evaluate deep architectures on public datasets.
4. Explain and discuss the theoretical underpinnings behind deep learning architectures and the methods in which they are trained.
5. Discuss how to apply deep learning methods to novel problems.
When the unit is taken as the MAJOR 20 credit variant, students will also be able to:
6. Replicate published experiments from published academic papers.
7. Discuss success and failures of current architectures.
8. Build upon published work in the area using newer and current approaches.
The unit includes lectures and a series of applied labs that allow students to learn the practical aspects of training deep learning methods. These labs include contact time with Teaching Assistants but can also be completed asynchronously and students are encouraged to explore the wider literature. Given the cutting-edge nature of the content of this unit, this gives the best combination of students dictating their own learning with support from experts within the field. If taken as a MAJOR option, the unit also provides weekly coursework support sessions.
Tasks which help you learn and prepare you for summative tasks (formative):
Teaching will take place over Weeks 1-7, with coursework support in weeks 9-11 and for students assessed by examination, consolidation and revision sessions in Weeks 12.
The unit begins with a series of lectures which covers the theoretical aspects of deep learning and its history with practical labs running alongside them once the basics have been covered. These practical labs have been designed to explore the theoretical content of the unit that is both beneficial for both the exam and coursework unit content. Labs include opportunities for students to discuss and check their progress on the learning outcomes of the unit. For students taking the MAJOR unit, code developed during the labs will be a foundation of their final summative assessment.
Tasks which count towards your unit mark (summative)
For students taking this unit with the Advanced Topics in Computer Science (MINOR) examination unit, it will contribute 50% towards the 20cp Advanced Topics in Computer Science exam, (equivalent to 1 hour of exam time) that will be sat during the winter examination period. This closed-book exam will assess Learning Outcomes 1, 2, 3, 4, and 5.
For students taking this unit as a MAJOR variant, there will be two elements of assessment:
- A mid-term in-class test that will assess Learning Outcomes 1, 4, 5, 7 (worth 30% of the unit) - An end-of-term coursework (programming exercise + written report), (taking place during weeks 9-11) that will assess Learning Outcomes 2, 3 and 6 (worth 70% of the unit)When assessment does not go to plan
Students will retake relevant assessments in a like-for-like fashion in accordance with the University rules and regulations.
If this unit has a Resource List, you will normally find a link to it in the Blackboard area for the unit. Sometimes there will be a separate link for each weekly topic.
If you are unable to access a list through Blackboard, you can also find it via the Resource Lists homepage. Search for the list by the unit name or code (e.g. COMSM0045).
How much time the unit requires
Each credit equates to 10 hours of total student input. For example a 20 credit unit will take you 200 hours
of study to complete. Your total learning time is made up of contact time, directed learning tasks,
independent learning and assessment activity.
See the University Workload statement relating to this unit for more information.
Assessment
The Board of Examiners will consider all cases where students have failed or not completed the assessments required for credit.
The Board considers each student's outcomes across all the units which contribute to each year's programme of study. For appropriate assessments, if you have self-certificated your absence, you will normally be required to complete it the next time it runs (for assessments at the end of TB1 and TB2 this is usually in the next re-assessment period).
The Board of Examiners will take into account any exceptional circumstances and operates
within the Regulations and Code of Practice for Taught Programmes.
