Example event in a LArTPC.
I am a PhD research student working on the development of optimised automated event reconstruction for the Deep Underground Neutrino Experiment (DUNE) using machine learning and a multi-algorithm approach to pattern recognition. I am a part of the Elementary Particle Physics group at the University of Warwick and my supervisor is Professor John Marshall.
DUNE is an international experiment for neutrino science located in the United States. It will deploy a Liquid-Argon Time Projection Chamber (LArTPC) as its detector. LArTPC’s give photographic quality images of the charged particles produced in neutrino interactions. Pattern recognition on these images is a crucial contribution to DUNE. Whilst the human eye can usually pick out the key features, it is a significant challenge to develop an automated, algorithmic solution. This is the single step in which LArTPC images are examined in detail, so it is vital that information in the images is fully extracted.
The Pandora framework champions a multi-algorithm approach to analysing LArTPC images. Many tens of algorithms carefully build up a picture of events and collectively give a strong reconstruction. This project will involve the development of novel pattern-recognition algorithms in this framework.
The early parts of the project will involve understanding the current performance of the Pandora pattern recognition for DUNE, finding any weaknesses and designing algorithms to address any issues with specific topologies. The project will then increasingly focus on the use of machine-learning opportunities to drive the decisions made by the algorithms. Ultimately a physics analysis will be developed, using detailed knowledge of the pattern recognition outputs.
Year 1 Library Demonstrator in Term 1.
I did my Undergraduate studies at the University of Warwick.
Department of Physics
University of Warwick
West Midlands, UK