EPSRC Industrial CASE PhD Studentship: Deep Learning by Cascading Radial Basis Function (RBF) Networks

Aston University

(3 years)

Do you want to learn about the first deep learning architecture that does not suffer from the problem of adversarial examples, has its own internal mechanism to have low confidence when it's predictions are weak, and also have extra financial support from industry (Thales UK)?

Applications are invited for a three-year eCASE Postgraduate studentship leading to a PhD. The studentship is financially supported by the Engineering and Physical Sciences Research Council (EPSRC) and Thales Research & Technology. The student will be working within the Systems Analytics Research Institute at Aston University.  The successful applicant will join a small, world-leading team developing advanced machine learning and visual analytics techniques appropriate for various problem domains such as image and text analysis, functional and network (graph) problems, social media community detection, anomaly detection, etc.

The position is available to start in January 2018 (or later by agreement) and is 3 years in duration

Financial Support

This studentship includes a fee bursary to cover the home/EU fees rate, plus a maintenance allowance of £14,296 in 2016/17 (subject to eligibility) plus a £3,000 per year award from the collaborating company. This application is only available to Home/EU students.

Background of the Project

Current state-of-the-art approaches to deep learning for big data analysis suffer from the problem of “adversarial examples” --- input patterns of one class which are imperceptible perturbations of another pattern and yet are assigned to completely different classes by the deep learning architecture. The project will investigate a new type of deep learning architecture invented at Aston University known as a cascading radial basis function network, which does not suffer from the same problems as current deep learning approaches such as adversarial examples yet has impressive learning capability. The student will be researching improving the algorithms, implementing the architecture on various real world problem domains, and developing new understanding on the capabilities of this alternative machine learning approach.  

Person Specification

The successful applicant should have a first class or upper second class honours degree or equivalent qualification in theoretical physics, applied mathematics, computational mathematics, electronic engineering systems, or similar. Applicants should fulfil the eligibility criteria for EPSRC funding through UK nationality and/or residency status (see http://www.epsrc.ac.uk/).

For informal enquiries about this project and other opportunities within the Systems Analytics Research Institute contact Dr Yulan He (y.he9@aston.ac.uk) or Professor David Lowe (d.lowe@aston.ac.uk).

If you require further information about the application process please contact the Postgraduate Admissions team at seasres@aston.ac.uk

The online application form, reference forms and details of entry requirements, including English language are available at http://www1.aston.ac.uk/eas/research/prospective-research-students/how-to-apply/ Applications should also be accompanied by a brief review  (2 pages) of relevant research literature appropriate for this project area.

Details of how to write your project proposal are also included in the How to Apply section.

Closing Date: 30th November 2017

Share this PhD
     
  Share by Email   Print this job   More sharing options
We value your feedback on the quality of our adverts. If you have a comment to make about the overall quality of this advert, or its categorisation then please send us your feedback
Advert information

Type / Role:

PhD

Location(s):

Midlands of England