Open Course Materials

ATML: Advanced Topics in Machine Learning

mcorey — Mon, 17 Mar 2025 09:43:52 +0000

ATML: Advanced Topics in Machine Learning mcorey Mon, 17/03/2025 - 09:43

Welcome

Welcome to the course webpage for Advanced Topics in Machine Learning. This course is designed for students who aspire to become technical experts, pursue research, and innovate in the field of machine learning.

This page will be updated with further information about the course, including the syllabus, schedule, and other resources.

Learning outcomes

Identify how an aspect of an advanced machine learning topic applies to a given applied problem.
Derive mathematical details of machine learning methods in the topic area.
Critically compare and contrast alternative choices or variants of methods or approaches in the area.
Create accessible and useful explanations of the workings and failure modes of machine learning methods, including appropriate mathematical and implementation detail.
Identify the ethical and societal implications, including both benefits and risks, of the deployment of machine learning methods in the area.

Lecture recordings

All lecture recordings should be accessed via Learn; you will need to log in using your EASE account. (Learn provides you with access to any lecture recordings available for this course. You will need to select the "lecture recording" link once, before you can access any direct links to a lecture recording.)

Topic tracks

Three tracks will be offered in the course in 2026. It is expected that students will follow two out of the three tracks.

Geometric learning (track page)

Topics: graph neural networks, models on non-Euclidean domains (e.g., manifolds such as spheres), symmetries, invariances and equivariances.

Instructor: Viacheslav Borovitskiy

Deep generative modelling (track page)

Topics: learning and inference in families of neural probabilistic models used for high-dimensional data: deep latent variable models (e.g., autoencoders), generative adversarial models, normalising flows, diffusion models. Principles of representation learning, generalisation, model assessment and deployment considerations.

Instructor: Nikolay Malkin

Optimisation in machine learning (track page)

Topics: non-convex optimisation and neural networks, loss landscapes in high dimensions, algorithmic stability, overfitting and overparameterisation in neural networks.

Instructor: Rik Sarkar

License

D2AIR: Dependable and Deployable Artificial Intelligence for Robotics 2

aburford — Thu, 30 Jan 2025 10:03:07 +0000

D2AIR: Dependable and Deployable Artificial Intelligence for Robotics 2 aburford Thu, 30/01/2025 - 10:03

This course is for students in the D2AIR CDT, and will be taught in two parts, the first by Prof. Barbara Webb (B.Webb@ed.ac.uk) and the second by Prof. Subramanian (Ram) Ramamoorthy (S.Ramamoorthy@ed.ac.uk).

The course has the following aims:

Introduce the students to different approaches to system design, concentrating on issues arising in the design of closed-loop robotics systems operating in human-centred and dynamic field environments.
Ensure students understand how responsible research and innovation issues can and should be incorporated at the stages of concept development and system design, with a specific focus on issues arising in RAS domains.
Make students aware of methodologies for measurement, testing and evaluation of the performance of robotics and autonomous systems (RAS) in field environments.
Through case studies and real-world examples, expose students to dependability and deployabiity issues arising in the field. This includes discussion and reflection on how design choices, integration and testing methodologies help or hinder safe field deployments of RAS technologies.

License

MDI: Masters Dissertation (Design Informatics)

mcorey — Fri, 13 Dec 2024 14:34:58 +0000

MDI: Masters Dissertation (Design Informatics) mcorey Fri, 13/12/2024 - 14:34

Please find all materials for this course here.

License

AFDS: Algorithmic Foundations of Data Science

flittlet — Mon, 04 Nov 2024 14:32:15 +0000

AFDS: Algorithmic Foundations of Data Science flittlet Mon, 04/11/2024 - 14:32

Welcome to Algorithmic Foundations of Data Science!

Learning Outcomes

On completion of this course, the student will be able to:

demonstrate familiarity with fundamentals for processing massive datasets.
describe and compare the various algorithmic design techniques covered in the syllabus to process massive datasets
apply the learned techniques to design efficient algorithms for massive datasets
apply basic knowledge in linear algebra and probability theory to prove the efficiency of the designed algorithm
use an appropriate software to solve certain algorithmic problems for a given dataset

Course Outline

The course aims to introduce algorithmic techniques that form the foundations of processing and analysing massive datasets of various forms. In particular, the course discusses how to pre-process massive datasets, efficiently store massive datasets, design fast algorithms for massive datasets, and analyse the performance of designed algorithms. Through various examples and the coursework, the students will see applications of the topics discussed in class in other areas of computer science, e.g., machine learning, and network science.

License

CV: Computer Vision

flittlet — Mon, 04 Nov 2024 11:37:16 +0000

CV: Computer Vision flittlet Mon, 04/11/2024 - 11:37

Welcome to Computer Vision!

Computer Vision is the field of study that teaches computers how to 'see'. This means, how to go from the pixels in an image to the information that a human can describe when they see a picture, much like self-driving cars, autonomous robots, or social media apps that recommend images or videos based on your preferences. This course is an in-depth introduction to the field of Computer Vision.

The course is structured around different problems in computer vision, such as object recognition and video classification, and covers both classical and deep learning approaches.

The course can be taken without any prior knowledge of computer vision or deep learning, but it does assume some familiarity with machine learning concepts, and relevant mathematics and programming skills (see details under "Other Requirements"). The course delivers both theoretical and practical knowledge, and by the end you should be able to understand, design, and implement computer vision techniques for many real-world problems.

Learning Outcome

On completion of this course, the student will be able to:

1. Define and explain principles underpinning computer vision methods
2. Describe current vision problem settings and their current solutions
3. Implement, train and debug computer vision models
4. Design, explain, analyse, and compare the behaviour of computer vision models under different settings
5. Identify social and ethical implications of computer vision methods in the real world

Learning Activities

The course will be taught as a combination of:
- Live lectures.
- Tutorials to develop your ability to solve vision problems from a theoretical perspective.
- Lab sessions to develop practical skills. The coursework will be structured as a series of small non-assessed practice sessions, which will build up the skills for the assessed mini-project at the end of the course.

License

THF: The Human Factor

flittlet — Mon, 28 Oct 2024 14:33:21 +0000

THF: The Human Factor flittlet Mon, 28/10/2024 - 14:33

Welcome to The Human Factor.

Course Overview

The Human Factor (THF) is a 10-credit elective course taught at the University of Edinburgh in Semester 2. The teaching team consists of lecturers Tara Capel (course organiser) and Susan Lechelt, teaching assistant (TA) Srravya Chandhiramowuli, and coursework markers Jenny Long and Diane Lac.

The course has no specific requirements for pre-requisite knowledge before the start of the course. Some of the material will be familiar to some people, especially those with a background in human-computer interaction, human factors, or user experience.

The course will be taught using a flipped classroom – before class you will work through materials. During class, we will work on activities designed to review the material and deepen your learning.

The content for each week will be available on OpenCourse for THF by Wednesday of the previous week. This means that we will start with the materials for Week 1, materials for Week 2 will be published by Wednesday of Week 1, and so on.

What is the course about?

“If the user can’t use it, then it doesn’t work at all” (Susan Dray).

This is an introduction to the fields of Human Factors and User Experience with an emphasis on developing practical skills that are grounded in a strong knowledge of theory.

When technical systems that have been crafted in years of painstaking work fail in practice, more often than not this is due to a lack of fit between the complex system, the people who interact with it, and the contexts in which it is used. In the best case, failure is just annoying, in the worst case, it costs lives.

In this course, we will look at the art and craft of building technical systems that people can actually use successfully. To this end, we will draw on relevant findings from anthropology; behavioural, cognitive and social psychology; human-computer interaction; and sociology.

Learning Outcomes

On successful completion of this course, you will be able to:

Understand how relevant aspects of context affect the interaction between people and technical systems, with a particular emphasis on anthropometric, behavioural, cognitive, and social factors (ABCS)
Assess the usability of a technological artefact, including both hardware and software, given a particular context of use
Integrate user experience and human factors into the process of designing or improving a technological artefact
Ensure that systems are resilient and learn from user errors

Where will we meet?

Lectures

The lectures for THF are on Tuesday 16:10-17:00. We will start promptly, as the 10-past the hour start time exists to ensure you have time to get from any prior classes. The lectures will be in person in Lecture Theatre G.03 50 George Square.

Q&A Session

The Q&A session for THF are on Thursday 16:10-17:00. These sessions are designed for students to drop in and ask questions and discuss any of the course content or get feedback on coursework.

Assignments

THF is marked using the University of Edinburgh’s Common Marking Scheme. This means that a grade of 80 and above is outstanding, a grand of 70-79 is excellent, a grade of 60-69 is very good, and a grade of 50-59 is a pass. Below is an overview of the coursework assignments – please see the detailed coursework assignment brief for further information.

CW1: 50%

Usability and User Experience Evaluation Report: The aim of this assessment is to evaluate the usability and user experience of an existing technology and generate suggestions for improving the design with particular emphasis on human factors.

CW2: 40%

Group Presentation: The aim of this assessment is to create a series of technology designs to address the issues found in the usability and user experience evaluation report and prototype the design suggestions proposed. This will be presented via a group presentation in class.

CW3: 10%

Individual Reflection: The aim of this assessment is to critically reflect on the design process engaged in during the course and the application of human factors during this process.

Course Readings

This course has no mandatory readings, only recommended readings:

Ritter, Frank E.; Baxter, Gordon D; Churchill, Elizabeth F. (2014): Foundations for Designing User Centred Systems. Springer (main textbook) Tullis, Tom; Albert, Bill (2013): Measuring the User Experience. 2nd edition. Morgan Kaufman.
Cooper, Alan; Reimann, Robert; Cronin, David; Noessel Christopher (2014): About Face: The Essentials of Interaction Design 4th edition, Wiley Preece / Sharp / Rogers: Interaction Design. 4th Edition. Wiley.

License

MLS: Machine Learning Systems

flittlet — Mon, 14 Oct 2024 14:36:41 +0000

MLS: Machine Learning Systems flittlet Mon, 14/10/2024 - 15:36

Course Contacts

Your lecturer for this course is Luo Mai (luo.mai@ed.ac.uk)

Course Introduction

The course on 'Machine Learning Systems' introduces the design of such systems and highlights their application in the hands-on experience of large-scale AI infrastructure. Students will acquire the skills necessary to analyse and implement (i) systems that retrieve large-scale data and (ii) systems that train and deploy large-scale machine learning models.

Learning Outcomes

On completion of this course, the student will be able to:

Understand different types of data, queries, workflows, and architectures of machine learning systems. Demonstrate the appropriate choice and use of particular data structures, and architectures.
Construct, analyse and profile implementation to given machine learning systems and iteratively improve the performance of those systems.
Compare and evaluate different systems and suggest/synthesise an appropriate system adoption solution.
Present the system solutions and engage in professional dialogue with peers to improve their solutions.
Reflect on the wider quality and security issues of data and machine learning models when discussing with specialist practitioners.

Course Schedule

Please find the schedule and teaching slides on Learn.

Coursework

We have open-sourced all teaching materials related to our ML system coursework on GitHub:

https://github.com/ed-aisys/edin-mls-25-spring

License

CCN: Computational Cognitive Neuroscience

flittlet — Mon, 14 Oct 2024 12:55:36 +0000

CCN: Computational Cognitive Neuroscience flittlet Mon, 14/10/2024 - 13:55

Welcome to Computational Cognitive Neuroscience

Course Contacts

The lecturer for this course will be Peggy Seriès.

TA: Lars Werne

Course Content

The title of this course could really be "Computational Cognitive Neuroscience and Computational Psychiatry".

In this course we study how computations carried out by the nervous system lead to cognition, in particular perception, learning, and decision-making. we incorporate data from neurobiology and behavioral experiments, simulate certain aspects of it, and try to formulate theories about the brain. A domain of application that is strongly emphasized is the field of mental illness. This course can be seen as an introduction to computational psychiatry.

Apart from learning about the brain, healthy cognition and mental illness, you will also learn about numerical modelling of differential equations, probabilistic models, reinforcement learning models applied to human learning, as well as current research and pitfalls in modelling cognition and mental disorders.

For whom is this course?
This course should appeal to students who are interested in models of computation in human and animal brains, as well as mental illness. It complements other courses in the cognitive sciences, offering a more biological perspective.

For whom is it not?
The topics discussed in the course have inspired many machine learning solutions to real-life problems, however, we shall hardly discuss those. It should also be noted that, although interest is growing, the course has limited direct practical applicability outside academic research.

Prerequisites

No prior biology/neuroscience/psychiatry knowledge is required. The course was built assuming some background in computer science or related quantitative field, in particular familiarity with coding. We use a small subset of not very advanced math and machine learning in the lectures. Keywords: linear differential equations, Bayesian inference models, model fitting and model comparison. Please make sure you have experience with these concepts before enrolling in the course.

The course "Computational Neuroscience" (CNS) is highly recommended.

In the labs, we use Python (or Matlab if you prefer but the tutors will be using Python).
Basics of Python is thus required. Prior experience with programming is required.

Learning Outcomes

On successful completion of this course, you should be able to:

1. describe current computational theories of the brain and mental illness

2. read, understand, and have a critical opinion on scientific articles related to computational cognitive neuroscience and computational psychiatry

3. write and analyse simple computational models related to brain function in Python or MATLAB

4. write scientific reports on topics related to computational cognitive neuroscience

Course Outline

- Overview of computational neuroscience basics (models of neurons and networks)
- Reinforcement learning models for computational neuroscience
- Bayesian models for computational neuroscience (The Bayesian Brain)
- Computational modelling of behavioural data
- Models of decision-making
- Application to individual differences (e.g., autism) and mental disorders (e.g.,schizophrenia, addiction, and depressionm anxiety): introduction to Computational Psychiatry

License

SCM: Seminar in Cognitive Modelling

flittlet — Mon, 30 Sep 2024 14:49:30 +0000

SCM: Seminar in Cognitive Modelling flittlet Mon, 30/09/2024 - 15:49

Please find all materials for this course here.

License

PBI: Programming for Biomedical Informatics

flittlet — Tue, 17 Sep 2024 14:47:28 +0000

PBI: Programming for Biomedical Informatics flittlet Tue, 17/09/2024 - 15:47

Please find all materials for this course here.

License