Undergraduate Course: Data Analysis and Machine Learning in Physics (PHYS11074)
Course Outline
| School | School of Physics and Astronomy |
College | College of Science and Engineering |
| Credit level (Normal year taken) | SCQF Level 11 (Year 5 Undergraduate) |
Availability | Available to all students |
| SCQF Credits | 20 |
ECTS Credits | 10 |
| Summary | This course develops core skills related to data analysis using deep learning. After a recap of Python programming tools and statistical methods, the students learn about neural networks and frequently used advanced deep learning techniques, including generative models and transformers. These techniques are applied to current physics research. |
| Course description |
This two-semester course combines lectures and lab workshops in which students carry out practical programming exercises. The course assessment is by project work, in which students deploy the taught techniques to physics research problems. The first part of the course consists of a recap of scientific programming in Python and of statistical methods. Following this, the students develop a thorough understanding of the mathematical background of fully connected neural networks, which are the building block of more advanced algorithms. The second part provides an overview of frequently used deep learning techniques including generative models, graph neural networks, convolutional networks, and transformers. Some space in the curriculum is reserved for latest topics in this fast-moving field. On completion of the course, students should be able to deploy key statistical and machine learning techniques, use these to develop solutions to research questions, and to critically judge the performance and suitability of these methods.
Students will attend a lecture each week, and their primary engagement with the course will be through weekly workshops where the lecture content is developed into a set of exercises. This will be an opportunity to discuss with peers, with TAs, and with the lecturers, and to receive feedback on work in progress. The majority of the assessment will be through 4 projects, spaced regularly throughout the year, where students will tackle a specific research question, submit a short report, and discuss results with academics in an interview. Three of these projects will be assigned by the lecturers, and the final project will be on a topic of the students¿ choice, producing presentation slides and responding to academics¿ questions.
|
Entry Requirements (not applicable to Visiting Students)
| Pre-requisites |
|
Co-requisites | |
| Prohibited Combinations | |
Other requirements | None |
Information for Visiting Students
| Pre-requisites | For undergraduates:
Modern Physics (PHYS08045)
AND
( Fourier Analysis and Statistics (PHYS09055) OR Probability (MATH08066) )
AND
( Computer Modelling (PHYS09057) OR Numerical Recipes (PHYS10090) ) |
Course Delivery Information
|
| Academic year 2026/27, Available to all students (SV1)
|
Quota: 150 |
| Course Start |
Full Year |
Timetable |
Timetable |
| Learning and Teaching activities (Further Info) |
Total Hours:
200
(
Lecture Hours 20,
Supervised Practical/Workshop/Studio Hours 60,
Programme Level Learning and Teaching Hours 4,
Directed Learning and Independent Learning Hours
116 )
|
| Assessment (Further Info) |
Written Exam
0 %,
Coursework
100 %,
Practical Exam
0 %
|
| Additional Information (Assessment) |
- 20% engagement marking, assessed during workshops, simply requires discussion with TAs on a relevant topic
- 3x20% assigned projects, with short report submission and follow-up interview with academics
- 20% self-directed project, with presentation slides submitted and follow-up interview with academics
|
| Feedback |
- Students to discuss their first project submission in workshop session before the deadline, and receive feedback about whether they are meeting expectations
- Students to present their final research project to their peers, and receive (moderated) peer feedback
- Project hand-ins and interview feedback to be returned within 15 working days of submission.
|
| No Exam Information |
Learning Outcomes
On completion of this course, the student will be able to:
- Deploy key elements of statistical learning ¿ tools for modeling and understanding complex datasets
- Develop deep learning solutions to physics data analysis problems. Evaluate and optimise their performance.
- Critically judge which deep learning techniques should be deployed to contemporary physics research problems, and understand relevant systematic uncertainties.
- Demonstrate big-picture understanding of the current state of machine learning application in Physics research
|
Reading List
- The Elements of Statistical Learning, by Hastie; Tibshirani, Friedman
- Statistical Methods for Data Analysis in Particle Physics; by Luca Lista
- Hands-On Machine Learning with Scikit-Learn and TensorFlow; by Aurélien Géron
|
Additional Information
| Graduate Attributes and Skills |
Not entered |
| Keywords | Not entered |
Contacts
| Course organiser | Dr Liza Mijovic
Tel: (0131 6)50 6771
Email: Liza.Mijovic@ed.ac.uk |
Course secretary | Mrs Gillian MacDonald
Tel: (0131 6)51 7525
Email: gillian.macdonald@ed.ac.uk |
|
|
University Homepage