Academic supervisor: Prof Coralia Cartis (DPhil in Mathematics); IBM co-supervisor: Prof Lior Horesh, Dr Soumyadip Ghosh, Dr Songtao Lu
Optimization problems, of huge scale, form the modelling and numerical core of machine learning and statistical methodologies. A grand challenge in this area is the need to augment stochastic gradient optimization methods with inexact second-order derivative information, so as to obtain more efficient methods especially in the nonconvex case of deep learning. In this project, we will investigate ways to approximate second-order information in the finite-sum structure of ML optimization problems.
DPhil in Computational Discovery
The DPhil in Computational Discovery is a multidisciplinary programme spanning projects in Advanced Molecular Simulations, Machine Learning and Quantum Computing to develop new tools and methodologies for life sciences discovery.
This innovative course has been developed in close partnership between Oxford University and IBM Research. Each research project has been co-developed by Oxford academics working with IBM scientists. Students will have a named IBM supervisor/s and many opportunities for collaboration with IBM throughout the studentship.
Applicants who are offered places on the DPhil in Computational Discovery will be fully funded. As a minimum, the funding package will include fees at the Home/EU rate, and a stipend for UK students only at the standard Research Council rate (currently £15,009 pa) for the duration of fee liability (four years).
Applications must be received by 12 midday (UK time) on Friday 24 January 2020.