Postdoctoral Research Associate in Deep Learning

Location: Mathematical Institute, Andrew Wiles Building, Radcliffe Observatory Quarter, Woodstock Road, Oxford, OX2 6GG

We are currently inviting applications for a Postdoctoral Research Associate to work with Professor Justin Sirignano at the Mathematical Institute, University of Oxford. This is a three-year, fixed-term position, funded by a research grant from the Engineering and Physical Sciences Research Council (EPSRC). The starting date of this position is flexible with an earliest start date of 01 March 2024. We particularly welcome applications from individuals who are able to start between 01 March and 01 October 2024.

The successful candidate will be part of a research group funded by a joint NSF and EPSRC grant to the University of Oxford and Boston University: “DMS-EPSRC: Asymptotic Analysis of Online Training Algorithms in Machine Learning: Recurrent, Graphical, and Deep Neural Networks”. The research will involve collaboration between the postdoctoral research associate, Prof. Justin Sirignano (University of Oxford), and Prof. Konstantinos Spiliopoulos (Boston University).

Neural network models in machine learning have achieved immense practical success over the past decade, revolutionizing fields such as image, text, and speech recognition. The training algorithms used for these complex machine learning problems -- although successful in practice -- are often ad hoc. Mathematical theory is yet to be established in many cases, and there is the potential to improve training algorithms and models via rigorous mathematical analysis.

The objective of this research project is to develop new mathematical theory for the training algorithms and neural network models used in several key areas of machine learning. Analysis will leverage methods from stochastic analysis and weak convergence theory to study the asymptotics of online, stochastic training algorithms and neural network models as the number of hidden units becomes large. The research project is highly interdisciplinary, integrating methods from probability, partial differential equations, stochastic analysis, optimization, and machine learning.

Please direct informal enquiries to the Recruitment Coordinator (email: vacancies@maths.ox.ac.uk), quoting vacancy reference 168225.

Applicants will be selected for interview purely based on their ability to satisfy the selection criteria as outlined in full in the job description. You will be required to upload a statement setting out how you meet the selection criteria, a curriculum vitae including full list of publications, a statement of research interests, and the contact details of two referees as part of your online application. (NOTE: Applicants are responsible for contacting their referees and making sure that their letters are received by the closing date).

Applications for this vacancy are to be made online. To apply for this vacancy and for further information, including a job description and selection criteria, please click on the 'Apply' button above.

Only applications received before 12.00 noon UK time on Monday 04 December can be considered.

Advert information