• The key research goal of Responsible AI is to develop new artificial intelligence and machine learning models that embed fairness, accountability, transparency, and trustworthiness into them for ensuring ethical outcomes and long-term public confidence in the deployment of automated systems.
• Rankings are based on CSRankings for Machine Learning. CSRankings is a metrics-based ranking of top computer science institutions around the world. Rankings are compiled from the number of computer science publications presented at the most prestigious publication venues.
The General Data Protection Regulation (GDPR) states data should be processed lawfully, fairly and transparently. With this in mind, BayesianGDPR project aims to integrate the legal non-discriminatory principles of GDPR into automated machine-learning systems in a transparent manner. It will do so by using a novel Bayesian approach to model all sources of uncertainty, and taking into account feedback from humans and future consequences of their outputs. BayesianGDPR will provide organisations that rely on machine learning technologies with concrete tools allowing them compliance with the non-discriminatory principles of GDPR and similar laws. The project's achievements will have an impact on computational law research and its integration into mainstream legal practice. It will also promote public confidence in machine learning systems.
Funded by European Research Council (ERC): 1 April 2020 - 31 March 2025 (€ 1,443,697).
Equipping ML models with ethical and legal constraints is a serious issue as without this the future of ML is at risk. In the UK, this is recognized by the House of Commons Science and Technology Committee, which has formed a Council of Data Ethics.
Building ML models with fairness, confidentiality, and transparency constraints is an active research area, and disjoint frameworks are available for addressing each constraint. However, how to put them all together is not obvious. Our long-term goal is to develop an ML framework with plug-and-play constraints that is able to handle any of the mentioned constraints, their combinations, and also new constraints that might be stipulated in the future.
Funded by Engineering and Physical Sciences Research Council (EPSRC): 1 October 2017 - 31 March 2019 (£100,675).
Surgo Foundation is launching the Surgo Machine Learning Initiative for Precision Public Health to explore the feasibility of applying causal machine learning methods to international development data. Surgo has formed a strong and diverse consortium of partners across the private and non-profit sectors including the Bill and Melinda Gates Foundation (BMGF), GNS Healthcare, the University of Manitoba, and the University of Sussex.
In its first proof-of-concept project, ML4PxP will begin by testing several potential causal machine learning approaches on reproductive, maternal, and child health data sets from Uttar Pradesh, India. Together, the consortium is innovating to determine whether and how such models can be applied to help solve big international development questions.
Funded by Surgo Foundation: 15 April 2017 - present.
Rapid urbanisation creates trade-offs between development, food security and poverty alleviation goals which are often ignored or invisible. Revealing and communicating the nature and scale of these trade-offs to policymakers is a key step towards achieving SDGs around urban sustainability and resilience.
Our project applies deep learning techniques to map peri-urban agriculture in Wuhan, China and explores ways of integrating multiple types of data through a web-based mapping and visualisation tool to support research and stakeholder engagement on urban sustainability policy. This is a cross-departmental project involving the Science Policy Research Unit (SPRU) at the University of Sussex.
Funded by Sussex Sustainability Research Programme: 1 April 2017 - 30 September 2018; British Academy: 18 November 2019 - 17 November 2021 (£244,847).
Tech giant Huawei uses machine learning algorithms to improve the image quality on millions of their smartphone devices. In addition to internal research, Huawei invited UK universities to compete at creating models which remove noise from the captured image.
A team of PAL PhD students entered the month-long competition. Using a novel approach to incorporate camera meta-data into the denoising process, they placed amongst the winners, taking home a cash prize plus an all-expenses paid trip to the company's headquarters in China.
The PAL laboratory was co-founded by Novi Quadrianto and Jeremy Reffin in 2017. We now have a team of 12 members consisting of faculty, research fellows, research associates and PhD students.