Date: Friday 12 May 2023 - 15:00 (Europe/London)
Speaker: Dr Leigh H. Whitehead , Senior Research Associate in the high energy physics group of the University of Cambridge
Abstract
Deep learning has become a major part of neutrino physics in the last decade. In this talk I will discuss the CNN-based algorithm, called the CVN, used to predict the flavour of neutrino interactions in the Deep Underground Neutrino Experiment (DUNE), a next generation long-baseline neutrino oscillation experiment. I will then present the results from a study of using transfer learning as a method to reduce the amount of simulated interactions required for training deep learning models for neutrino event classification.
The talk is now available on YouTube: https://youtu.be/rZVPPfaZjqw
Biography
Dr Leigh H. Whitehead is a Senior Research Associate in the high energy physics group of the University of Cambridge. He completed his PhD at the University of Warwick in 2012 after studying neutrinos from the T2K experiment for four years.
He then moved to UCL (2012 - 2016) as a Research Associate working on the MINOS+ and CHIPS experiments, before moving to CERN as a Research Fellow (2016-2019), focused on the ProtoDUNE and DUNE experiments. He joined Cambridge in 2019 and continues to work on DUNE and ProtoDUNE, with a particular focus of applying deep learning to different aspects of neutrino physics.
His research has taken him to many parts of the world, spending time living in Japan, the US and Switzerland.
