Rohan Teelock-Gaya

NEF CDT Student
Department of Engineering
University of Cambridge

Contact: rnt26@cam.ac.uk

Sponsor: AWE and EPSRC

Machine Learning of Nuclear Data

Nuclear reaction cross sections are determined using experimentation or theoretical models of reactions. Both methods have their own limitations: experiments are considered the gold standard, but are difficult to run for highly unstable or difficult to produce nuclei, known as "exotic" nuclei. Theoretical methods are known to be approximations of cross section behaviour, and are usually the only source of cross sections for exotic nuclei.

My current research focuses on using machine learning techniques to produce alternative predictions for exotic cross sections. This involves training models on evaluated cross sections from nuclear libraries such as ENDF/B-VIII, as well as other nuclear data which describe the target nuclei and the reactions taking place. So far, model predictions seem to be quite promising.

Future Plans

A career in nuclear data, perhaps as a nuclear data evaluator, is my end goal.

Achievements

Conference paper accepted by PHYSOR 2024. 
Awarded course scholarship for MPhil in nuclear energy at Cambridge.
J.D. Lewins prize for best MPhil dissertation.

Highlights

Travelling for CDT events