Department of Computing researchers selected to present research in Parliament

by

Miten Mistry at the Centre for Process Systems Engineering Industrial Consortium Meeting

Miten Mistry presents his research

Miten Mistry and Simon Olofsson have been selected for the 2019 STEM for Britain exhibition at the House of Commons where they will present to MPs.

Department of Computing PhD students Miten Mistry and Simon Olofsson have been selected to compete in the STEM for Britain poster exhibition on Wednesday, 13 March 2019. During the event, Miten and Simon will present their PhD research at the House of Commons to Members of both Houses of Parliament at Westminster. Simon and Miten look forward to discussing their research with their local MPs and members of the Parliamentary & Scientific Committee.

Miten Mistry, a PhD student in the EPSRC HiPEDS Centre for Doctoral Training, will present his work on Optimising over machine learning tree-based models in the Mathematical Sciences Session. In this work, Miten integrates a machine learning function, which is already computed from data, into a larger decision-making problem. Dr Robert Lee, a Research Scientist in Statistics and Machine Learning at BASF who collaborates on the project, says that "Miten’s mix of skills allowed him to take advantage of the special structure of GBT models; the success of our project is due in large part to his strong grasp of complex concepts across different fields and the creativity with which he combined different approaches."

Simon Olofsson, a PhD student associated with the Horizon 2020 ModLife ITN, will present his work on Design of Experiments for Model Discrimination using Gaussian Process Surrogate Models in the Engineering Sciences Session. Dr Marc Deisenroth, a Senior Lecturer in the Imperial Department of Computing, comments that "Simon’s major theoretical contribution to design of experiments for model discrimination is to replace the original models with Gaussian process surrogate models. By using Gaussian process surrogates, Simon makes the entire model discrimination process more accessible." Simon also makes his work publicly available, e.g. the open-source Python library GPdoemd.

STEM for Britain consists of poster exhibitions and judging sessions. The competition has approximately 500 applications each year, 35% of which are selected to present in Parliament. Congratulations, Miten and Simon!

Supporters

Reporter

Ruth Misener

Ruth Misener
Department of Computing

Click to expand or contract

Contact details

Tel: +44 (0)20 7594 8315
Email: r.misener@imperial.ac.uk

Show all stories by this author

Tags:

Outreach, Students, Research
See more tags