Computer Science PhD Studentship in Machine Learning and Optimisation: Ensemble learning of multiple objectives University of Exeter


University of Exeter – Computer Science

Primary supervisor: Dr Jonathan Fieldsend

Many machine learning problems involve a trade-off between more than one objective. For example, different measures of accuracy/quality, computational cost, solution robustness, etc. Multi-objective evolutionary algorithms have been used to find the optimal trade-offs between these objectives.

Ensemble learning allows the variance in predictions to be reduced by combining predictions from an ensemble of learners. Members of the ensemble are made diverse by giving them different architectures or by training on different views of the data. Usually the ensemble’s members are specialised for one particular objective, but for multi-objective learning and prediction, we seek to learn from ensembles of learners that have diverse skill in predicting the objectives. This project will investigate the use of ensembles of multi-objective learners for learning in big data applications.

Contact for Informal Enquiries:  Dr Jonathan Fieldsend ([email protected]; tel: +44 1392 722090), or Prof. Richard Everson ([email protected]; tel: +44 1392 724065)

Application Criteria: Applicants should have or expect to achieve at least a 2:1 honours degree, or equivalent, in Computer Science, Mathematics, or an aligned subject. It would be beneficial for applicants to have had experience of one or both of: nature inspired computing and machine learning.

If English is not your first language you will need to meet the English language requirements and provide proof of proficiency. Click here for more information and a list of acceptable alternative tests.

Streatham Campus, University of Exeter, EX4 4QJ

Funding Maximum 3.5 year studentship: UK/EU tuition fees and an annual maintenance allowance at current Research Council rate of £14,296 per year

Apply
source: jobs.ac.uk

Leave a Reply