Research Assistant in Spatio-Temporal Inference for Networked Objects (Part Time, Fixed Term)
Department of Engineering
A position exists, for a part time (10 hours per week) Research Assistant in the Department of Engineering, to work with Prof. Simon Godsill and his team on the project: "SIGNetS - Signal and Information Gathering for Networked Surveillance".
The post holders will be located in Central Cambridge Cambridgeshire, UK.
This project aims at developing scalable Bayesian approaches able to solve complex multi-sensor, heterogeneous data problems, aimed at improving localisation, situational awareness and latent intent prediction. The work will develop new approximate inference paradigms based around such techniques as variational inference, Gaussian processes and Monte Carlo to solve problems with many sensors that have limited communication abilities and localised computational resources. Objects to be localised will typically be in coordinated network formations and part of the inference task is to determine the latent connectivities between these objects and to learn their intentionalities. Modelling for individual objects will include investigation of new non-Gaussian Levy process formulations that allow for highly erratic and evasive manoevres and require advanced computational methodologies to infer their behaviour.
https://www.jobs.cam.ac.uk/job/35765/