The SPHERE project (a Sensor Platform for HEalthcare in a Residential Environment), funded by EPSRC, has been developing a unique integrated platform of sensors to deploy in people’s homes to monitor their health and wellbeing during everyday life (
http://irc-sphere.ac.uk). Within this exciting interdisciplinary project we are looking for exceptional candidates to strengthen our data mining capability.
Successful candidates will work on robust, sustainable data integration and machine learning techniques for mining data from the diverse range of ambient, video, and on-body sensors deployed within this large project. They will build on the work done to date by the data mining and data fusion work team, as well as on a range of relevant machine learning and data mining expertise in the Bristol Intelligent Systems Laboratory.
For this post we are particularly looking for candidates with relevant experience in learning from varying degrees of supervision and/or leveraging synthetic data. The first topic anticipates cases in which a perfect ground truth cannot be obtained, and is related to learning from weak and noisy labels, modelling annotator variability, model-based learning, etc. The second topic is concerned with synthetically generated but realistic data matching different contexts and environments and builds upon current machine learning research such as transfer learning and generative adversarial models. The ideal candidate will therefore have a strong research track record in machine learning and data mining with particular experience in some of these topics; please expand on this in your cover letter.
For informal enquires contact: Professor Peter Flach,
Peter...@bristol.ac.uk
Further details incl. how to apply:
http://www.bristol.ac.uk/jobs/find/details.html?nPostingID=37314&nPostingTargetID=125774
Closing date for applications: 14th of April, 2019