Applications are invited for a postdoctoral position investigating "Training Neural Networks using Nested Sampling with applications to Facial Recognition" at the Cavendish Laboratory (Department of Physics, University of Cambridge).
The successful candidate will join the Astrophysics research group, undertaking work led by Anthony Lasenby, Will Handley and Mike Hobson, and will apply the latest developments in Nested Sampling (encapsulated in PolyChord 2.0) to the research problem of training Bayesian Neural Networks (BNN). As a test case they will use a state-of-the-art facial recognition training corpus, provided by Anyvision, a world leader in machine learning, with interaction facilitated by PolyChord Ltd.
The post is for one year, and is ideal for a recently graduated PhD student who wishes to further their skills in machine learning and interaction with industry. Some background in MCMC techniques and machine learning is essential, with experience in using nested sampling packages and techniques (PolyChord, MultiNest, DyNesty, DNest) highly desirable. The role holder wil also be expected to interact with industrial collaborators on a regular basis.
The Astrophysics group has a long history of developing machine learning and nested sampling techniques and applying them in the context of Cosmology and Astrophysics, creating tools such as SkyNet, MultiNest and PolyChord. These techniques have also been applied to particle physics via GAMBIT and to geophysics problems as part of a Shell-funded postdoctoral scheme. This grant aims to transfer nested sampling techniques to the machine learning community as a whole.
Applications will be reviewed after the closing date and interviews conducted by mid-August 2018. The candidate would ideally start as soon as possible. The work will be funded by the STFC Impact Acceleration Account 2018 programme, provided by Cambridge Enterprise.
Fixed-term: The funds for this post are available for 12 months in the first instance.
Appointment at research associate is dependent on having a PhD (or equivalent experience), including those who have submitted but not yet received their PhD. Successful candidates who have not been awarded their PhD by the appointment date will be appointed as a Research Assistant at Grade 5 (£25,728 - £29,799 per annum). Upon award of the PhD the individual will be promoted to Research Associate, Grade 7 (£31,604 - £38,833 per annum).
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.
Please ensure that you upload your Curriculum Vitae (CV) including a list of publications and the contact details of at least two referees and a covering letter in the Upload section of the online application. If you upload any additional documents which have not been requested, we will not be able to consider these as part of your application.
If you have any questions about this vacancy or the application process, please contact Will Handley at firstname.lastname@example.org or Anthony Lasenby at email@example.com. Please quote reference KA16016 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.