Modelling Environmental Data Systems with Deep Learning Funding: Funded PhD Duration: 3 years (full-time) Application deadline: Wednesday 23 January 2019 How to apply About this project Apply deep learning and HPC to induce a joint and multi-dimensional data model from a range of available environmental data. The goal is to identify hidden patterns and generate inferences and predictions on water quality, nutrients and pollutants that can emerge only from a multi-dimensional perspective on the data rather than taking a restrictive approach. We will derive models about movement of chemicals and processes in the environment by integrating and processing data sets from our deployed sensors with openly available datasets including those from remote sensors (e.g. EU-Sentinel satellite images) and apply transfer learning approaches, i.e. transfer knowledge from data in one environmental domain to another to accelerate learning and seamlessly integrate the disparate origins of the data to develop an interconnected inference model. Project supervisors: Nina Dethlefs (Computer Science) Rodney Forster (Institute of Estuarine and Coastal Studies) Dan Parsons (Energy and Environment Institute) Supervisors We recommend contacting a supervisor to discuss your research proposal before making an application. Dr Nina Dethlefs Dr Rodney Forster Professor Daniel Parsons About Postgraduate Research at Hull The University's Postgraduate Training Scheme (PGTS) provides a range of generic and discipline-specific modules to support research students through their programme. Find out more The library has an exclusive lounge for postgraduate research students and a dedicated Skills Team to provide a wide range of study and research skills help. Find out more The Graduate School provides support to postgraduate research students. Offering skills development opportunities and dedicated facilities, the school is here to help you achieve your potential. Find out more Research at Hull tackles big challenges and makes an impact on lives globally, every day. Our current research portfolio spans everything from health to habitats, food to flooding and supply chains to slavery. Find out more Funding The studentships will include full fees and maintenance (£14,777 in 2018/19) for three years, depending on satisfactory progress. Entry requirements The candidate should have excellent programming skills and a degree in Computer Science or a related discipline. Knowledge or experience in one or more of the following areas is desirable: machine learning / deep learning, embedded systems, physical sciences, engineering or environmental sciences.