The Astrophysics Group of the Department of Physics, University of Cambridge, invites applications for two Postdoctoral Research Associates to work in the Bayesian Inference and Machine-Learning Research Team of Professor Hobson.
The team is involved in the development of Bayesian inference and machine-learning methods, and their application to the analysis of complicated astronomical data-sets. These techniques are also applicable to general inference problems beyond astrophysics. In particular, the multinational energy company Shell has funded these Research Associateships to investigate how such methods may be used in probabilistic seismic inversion problems and to build on existing work already completed as part of this collaboration.
The post holders will be expected to conduct independent research programmes and to participate actively in the scientific exploitation of seismic data analysis with Shell. In particular, the projects will focus on the use of nested sampling methods, embodied in the MultiNest software package, to estimate the Bayesian evidence for different physical models and explore the posterior probability of their parameters. The key aspects of interest are: efficiency; accuracy of the estimate of the posterior and evidence; capability to sample multi-modal distributions; and capability to sample a combination of continuous and categorical variables. Another key aspect of the projects will concern the use of machine-learning methods for building proxy/surrogate models for the likelihood function to increase the speed of the analysis. Two main projects are envisaged as follows. Project 1 will further develop a systematic approach for detecting microseismic events, consisting of forward simulation for elastic wave propagation, learning sparse seismic patterns using surrogate regression models, and Bayesian inference of the microseismic sources using nested sampling algorithms. In particular, the project will concentrate on testing alternative regression proxy models with a specific focus on deep learning, and application on real geophysical data. The extension of this approach to reflection seismic inversion and inference problems will also be investigated. Project 2 will focus on the application of statistical learning and inference methods to the cleaning of and event detection in seismic shot records from boreholes for the purpose of Vertical Seismic Profiling (VSP). This will require development of machine-learning methods to map a large input data set to an an equally large output data set, using primarily deep learning neural network techniques. Applicants should have a PhD in Physics, Mathematics, Engineering, Computer Science or another relevant scientific discipline. They should have a proven track record in independent research and scientific exploitation of data. The candidate must be familiar with Bayesian inference, machine-learning and data analysis in general. Experience with GPU programming is desirable. Applicants will also be expected to spend significant time (several weeks per year) at Shell Rijswijk to receive proper support and maintain the connection with the Shell researchers.
On appointment, the post-holders will be based at the Battcock Centre for Experimental Astrophysics at the Cavendish Laboratory, Department of Physics (J J Thomson Avenue, Cambridge CB3 0HE).
This post is funded until 31st Aug 2019 in the first instance.
To apply online for this vacancy, please click on the 'Apply' button below. This will route you to the University's Web Recruitment System, where you will need to register an account (if you have not already) and log in before completing the online application form.
If you have any questions about this vacancy please contact Professor Mike Hobson, email@example.com
For all administrative queries, please contact firstname.lastname@example.org
Please quote reference KA13795 on your application and in any correspondence about this vacancy.
The University values diversity and is committed to equality of opportunity.
The University has a responsibility to ensure that all employees are eligible to live and work in the UK.