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

Location: Streatham Campus, University of Exeter, EX4 4QJ

Primary Supervisor: Dr Jonathan Fieldsend (http://emps.exeter.ac.uk/computer-science/staff/jefields)

Secondary Supervisor: Professor Richard Everson (http://emps.exeter.ac.uk/computer-science/staff/reverson)

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: Professor Richard Everson [Email Address Removed]; tel: +44 1392 724065 or Dr Jonathan Fieldsend [Email Address Removed]; tel: +44 1392 722090

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. For more information and a list of acceptable alternative tests please visit http://www.exeter.ac.uk/postgraduate/apply/english/

The closing date for applicants is midnight on 15th January 2017