PhD Opportunities in the Institute for Mathematical Sciences, Imperial College London
We have two funded PhD studentships intended to commence in October 2009. These positions are in Professor David Hand's ALADDIN project (www.aladdinproject.org<http://www.aladdinproject.org/>) group. The ALADDIN project is a large multi-disciplinary research project funded by a BAE Systems and EPSRC strategic partnership, concerned with developing mechanisms, architectures, and techniques to deal with the dynamic and uncertain nature of distributed and decentralised complex systems. To be eligible for funding, applicants must be UK citizens, or EU citizens resident in the UK for the last three years.
Deadline for applications is 31 July.
To apply please email your CV, two references and a personal statement to [log in to unmask]<mailto:[log in to unmask]> or send them by post to Rusudan Svanidze, Department of Mathematics,180 Queen's Gate, London, SW7 2AZ. Short listed candidates will be asked to complete a postgraduate application form
1). Hybrid state-space models
State-space models provide an effective means for tracking dynamic phenomena. These models have wide applicability in monitoring and sensor networks that are a primary concern of the ALADDIN project. Real-life deployment of state-space models typically relies either on strong domain knowledge or computationally demanding learning algorithms. The approach is therefore hard to motivate in certain contexts, such as streaming data analysis, where a process of unknown dynamics needs to be monitored in real-time with limited computational resources.
This project will explore the utility of combining recently developed adaptive forgetting procedures with state-modelling to yield new state-space approaches that are efficient and adaptive. The new algorithms developed in the project will be probed theoretically, and tested in a variety of real applications.Applicants should have (at least) a good undergraduate qualification with substantial mathematical and statistics content, and very strong computing skills. The successful applicant will be expected to work jointly with other project members based at universities in Oxford, Southampton and Bristol. Please contact Niall Adams ([log in to unmask]<mailto:[log in to unmask]>) or Dimitris Tasoulis ([log in to unmask]<mailto:[log in to unmask]>) for more details.
2). Heterogeneous Data Fusion
Information obtained from different types of measurements of an organism can have very diverse characteristics. Measurements vary in terms of measurement scale, resolution, noise characteristics and much more. Furthermore, the integration of diverse types of data and information is important as it provides a much more specific and complete inferences about the organism, which cannot be achieved using one information source alone. In medicine, data fusion can ultimately lead to novel treatments and diagnoses. In disaster management, data fusion can literally save lives. The aim of this project is to develop novel mathematical techniques for multisensor data fusion in diverse applications such as functional magnetic resonance imaging and disaster management.
This project will research models, inference and application of the fusion of data and information. Applicants should have studied the theory of one or more of optimisation, multiple regression, survey sampling, or stochastic time series in a mathematical discipline at undergraduate level. It would be an advantage to have carried out project work that overlaps with these subject areas, and to have experience of computer programming. The candidate will be jointly supervised by the medical and statistical faculty (Department of Neuroscience and Mental Health and Mathematics Department, respectively). The successful applicant will be expected to work jointly with other project members based at universities in Oxford, Southampton and Bristol. Further project specific information can be obtained from I. Rezek <[log in to unmask]<mailto:[log in to unmask]>> or, in the second instance, from N. Adams <[log in to unmask]<mailto:[log in to unmask]>>.
|