Supervisor: Dr Charles Sutton, University of Edinburgh
Apply: http://www.ed.ac.uk/schools-departments/informatics/postgraduate/apply. Apply to the PhD programme in Adaptive and Neural Computation, mentioning this project
Deadline: Accepting applications as soon as possible, until post filled. Studentship to start September 2017
More and more industries are coming to rely on finding patterns and making predictions based on big data, but the first step in any big data effort is for a person to take a data set and try to develop a gut feeling of what’s in it. The goal of this project is to develop new tools based on modern machine learning and deep learning to identify patterns in big data sets that can then be shown to a person to help them develop their intuition about what types of knowledge is likely to be hidden in the data.
This project aims to bridge this gap by developing new deep generative models specifically for the purpose of aiding exploratory data analysis. The techniques used in this project may include probabilistic modelling, topic modelling, variational autoencoders, and generative adversarial networks.
This project is generously funded by a grant from Huawei, and there will be opportunities to interact with researchers and developers from Huawei. Additionally, this project will be aligned with the AIDA research project, on improving data analytics, at the Alan Turing Institute, the UK national national research institute for data science. There will be opportunities to collaborate with researchers on the AIDA project as well.
The project is suitable for a student with a top MSc or first-class bachelor’s degree in computer science, statistics, physics, or a related numerate discipline. Previous coursework or experience in machine learning is necessary. A good programming background will also be essential.
Here are a few recent papers from our research to provide a starting point for these ideas:
Autoencoding Variational Inference for Topic Models. Akash Srivastava and Charles Sutton. In International Conference on Learning Representations (ICLR). 2017. http://homepages.inf.ed.ac.uk/csutton/publications/avitm.pdf
A Bayesian Network Model for Interesting Itemsets. Jaroslav Fowkes and Charles Sutton. In European Conference on Machine Learning and Principles and Practice of Knowledge Discovery (ECML-PKDD). 2016. http://homepages.inf.ed.ac.uk/csutton/publications/pkdd2016-iim.pdf
The School of Informatics is one of the top-ranked departments of computer science in Europe, and one of the largest and best research centres in the world, with over 100 faculty and 400 PhD students and researchers. A national research assessment exercise (REF 2014) concluded that the School produced more world-leading and internationally excellent research in computer science than any other university in the UK. Our strength in machine learning, natural language processing, and the larger fields has been recognised by recent awards of the EPSRC Centre for Doctoral Training in Data Science (fully funds 50 PhD students over 8 years), and by being selected as one of only 5 UK universities to be a founding partner of the Alan Turing Institute.
The city of Edinburgh is a beautiful city of historic sandstone buildings, many of which are part of a UNESCO World Heritage Site, that provides a high quality of life — or, in the words of Alexander McCall Smith, “a city of shifting light, of changing skies, of sudden vistas. A city so beautiful it breaks the heart again and again.”
For more information, see http://www.inf.ed.ac.uk.
For more information on Dr Sutton or the research in his group, please see http://homepages.inf.ed.ac.uk/csutton/.
This is a fully funded studentship for UK, EU, and overseas students. The studentship will cover tuition, fees, and stipend for three years.
For informal enquiries about the studentship, please contact Dr Charles Sutton email@example.com.
Formal application must be through the School’s normal PhD application process: http://www.ed.ac.uk/schools-departments/informatics/postgraduate/apply. Select the Informatics: Institute for Adaptive and Neural Computation research area and mention that you are applying for this project.
For full consideration, please apply as soon as possible. We will consider applications as they are received, so that the successful applicant can begin in September 2017.