EPSRC gives £14 million to projects that take new approaches to Data Science
Supplementary content information
Five new research projects that take novel approaches to challenges in data science were announced today. The successful proposals were responding to a call (New Approaches to Data Science issued by the Engineering and Physical Sciences Research Council (EPSRC) which closed in January last year.
The research will be led by four universities, the University of Glasgow, Lancaster University, the University of Liverpool, and the University of Oxford.
Co-investigators will be drawn from the universities of Bristol, Cambridge, Liverpool, Oxford, and Warwick, Swansea University, the Natural Environment Research Council (NERC) Centre for Ecology and Hydrology, and Science and Technology Facilities Council's (STFC) Hartree Centre.
The projects will bring together statisticians, computer scientists and, in one case, environmental scientists, alongside an array of public and private sector partners and stakeholders. The projects will also look to partner with other bodies like The Alan Turing Institute.
They will explore new ways of applying machine learning methods and how to develop algorithms to deal with large data, but also using novel mathematics to obtain meaning from the shape of data as well how feedback loops affect data in real time.
The research will have relevance to a wide range of sectors including health sciences, security, transport, smart cities, finance, and the environment.
Universities and Science Minister, Sam Gyimah said:
We know the data we hold can change the way we live our lives and these important research projects will help us better understand the vast amount of data that is produced on a daily basis. Some of the best minds in our Research Councils and at higher education institutions will work collaboratively on these projects with industry and public bodies, helping to extract value and use data to assist with decision making.
Through our modern, ambitious Industrial Strategy and artificial intelligence Sector Deal and Grand Challenge, we will build on our reputation as a world-leader in this transformative technology, ensuring we make the very most of our data output and help build a Britain fit for the future.
In addition to EPSRC's funding the projects between them have attracted, further contributions from partners worth £3.7 million. The multidisciplinary research is involving companies and public bodies such as GSK, Unilever, Skyscanner, Dstl, the Met Office and Public Health England to name just a few.
The £14 million includes £500,000 from the Natural Environment Research Council (NERC) which is dedicated to the Data Science of the Natural Environment project at Lancaster.
Professor Philip Nelson, Chief Executive of the Engineering and Physical Sciences Research Council (EPSRC) said:
Data pervades almost every aspect of modern life. Collectively we are producing ever more data but we need the mathematical and systemic tools to deal with it, quickly and accurately to make productive use of it. These projects will help scientists and businesses make discoveries and better informed commercial decisions.
Closed-Loop Data Science for Complex, Computationally- and Data-Intensive Analytics (EP/R018634/1)
- Led by Professor Roderick Murray Smith, University of Glasgow
- EPSRC grant: £3.055 million
- Project Partner Leverage: £959,500
Progress in sensing, computational power, storage and analytic tools gives access to enormous amounts of complex data, which can inform us of better ways to manage our cities, run our companies or develop new medicines.
However, the 'elephant in the room' is that when we act on that data we change the world, potentially invalidating the older data.
To support human interactions with large and complex data sets, this project will look at the overlap between the challenge someone faces when coping with all the choices associated with booking a flight for a weekend away, and an expert running complex experiments in a laboratory.
The project will test the core ideas in a number of areas, including personalisation of hearing aids, analysis of cancer data, and adapting the computing resources for a major bank. The researchers will be working with a number of partners including JP Morgan and Skyscanner.
New Approaches to Bayesian Data Science: Tackling Challenges from the Health Sciences (EP/R018561/1)
- Led by Professor Paul Fearnhead, Lancaster University
- EPSRC grant: £2.95 million
- Project Partner Leverage: £193,000
Current Bayesian data science methods are not feasible for many modern, big-data, applications in the health sciences. Bayesian methods require integrating over uncertainty. Such high-dimensional integration carries a substantial computational overhead when compared to alternative, often optimization-based, data science methods.
So, while the motivation for Bayesian analysis is clear, this computational overhead means that, currently, implementing Bayesian approaches is often not feasible.
This grant will work on new Bayesian methods which are more applicable and usable by the health sciences. It will work with a number of health partners including GSK, AstraZeneca and Public Health England amongst others, using real data to test out their new methods.
Data Science of the Natural Environment (EP/R01860X)
- Led by Professor David Leslie, Lancaster University
- EPSRC grant: £2.6 million
- Project Partner Leverage: £542,000
This project brings together statisticians, computer scientists and environmental scientists alongside an array of public and private sector stakeholders to effect a step change in data culture in the environmental sciences. Modern advances in data science, such as machine learning, have been applied to areas such as eCommerce, smart cities, transport and health, but has yet to be fully deployed in one of the most important problems currently facing humanity - mitigating and adapting to climate change.
Specifically this grant focusses on three Grand Challenges in the area of environmental sciences - predicting ice sheet melt, modelling and mitigating poor air quality, managing land use for maximal societal benefit. The grant holders will create an integrated suite of novel data science tools - a modular platform which can be used by data scientists but also by environmental scientists and stakeholders without data science training.
This grant received £500,000 co-funding from NERC.
Big Hypotheses: A Fully Parallelised Bayesian Inference Solution (EP/R018537/1)
- Led by Professor Simon Maskell, University of Liverpool
- EPSRC grant: £2.5 million
- Project Partner Leverage: £1.2 million
A family of algorithms know as Markov Chain Monte Carlo provide impressive accuracy in extracting information from data but are incredibly computationally expensive. Users working across physics, chemistry, the life sciences, government and industry want to access the accuracy offered by MCMC but at a fraction of the current computational cost.
This grant proposes to use a new family of algorithms known as Sequential Monte Carlo. Essentially these algorithms are much easier to run in parallel than MCMC. This project not only involves the development of new algorithms but also new hardware including GPUs and Field Programmable Gate Arrays (FPGA) capable of running this algorithms in parallel. Project partners include Intel, IBM and NVidia who will help with the hardware side of the project.
Application driven Topological Data Analysis (EP/R018472/1)
- Led by Professor Ulrike Tillmann and Dr Heather Harrington, University of Oxford
- EPSRC grant: £2.8 million
- Project Partner Leverage: £799,500
Modern science and technology generates data at an unprecedented rate. A major challenge is that this data is often complex, high dimensional, may include temporal and/or spatial information. The "shape" (Topology) of the data can be important but it is difficult to extract and quantify it using standard machine learning or statistical techniques.
This project will be using pure mathematics to look at the shape of data to understand its properties. For example, an image of blood vessels near a tumour looks very different than an image of healthy blood vessels; statistics alone cannot quantify this shape because it is the shape that matters.
The team will study the shape of data, through the development of new mathematics and algorithms, and build on existing data science techniques. The researchers pose some interesting questions that the study of the shape of data might be able to discern:
- Can we detect a tumour by looking at the shape of images of blood vessels?
- Can we design new materials by looking at the shape of molecules using topology?
- How can we design such molecules?
- Can we detect anomalies in security data? And importantly, how can we accelerate algorithms to obtain topological characteristics of data in real time?
Researchers will work with domain-specific collaborators to develop the necessary theory, algorithms and statistics for implementation and interpretation. They propose crossing distinct branches of pure mathematics, applied mathematics, computer science, and statistics, to address ambitious and widely-applicable problems.
They will be combining theory, computation, and application, is to answer application specific questions.
Notes to editors:
The Engineering and Physical Sciences Research Council (EPSRC)
As the main funding agency for engineering and physical sciences research, our vision is for the UK to be the best place in the world to Research, Discover and Innovate.
By investing £800 million a year in research and postgraduate training, we are building the knowledge and skills base needed to address the scientific and technological challenges facing the nation. Our portfolio covers a vast range of fields from healthcare technologies to structural engineering, manufacturing to mathematics, advanced materials to chemistry. The research we fund has impact across all sectors. It provides a platform for future economic development in the UK and improvements for everyone's health, lifestyle and culture.
We work collectively with our partners and other Research Councils on issues of common concern via Research Councils UK.
Reference: PN 06-18