Application of Transfer Learning to Neutrino Interaction Classification

TL;DR

This paper explores transfer learning to improve neutrino interaction classification in liquid argon detectors, reducing training data needs and enhancing model performance compared to traditional methods.

Contribution

It demonstrates that transfer learning from generic image recognition models significantly improves neutrino event classification accuracy and bias reduction.

Findings

Transfer learning achieved an F1 score of 0.896 with 100,000 events.

Transfer-learned models show lower energy bias.

Transfer learning provides more balanced performance across interaction types.

Abstract

Training deep neural networks using simulations typically requires very large numbers of simulated events. This can be a large computational burden and a limitation in the performance of the deep learning algorithm when insufficient numbers of events can be produced. We investigate the use of transfer learning, where a set of simulated images are used to fine tune a model trained on generic image recognition tasks, to the specific use case of neutrino interaction classification in a liquid argon time projection chamber. A ResNet18, pre-trained on photographic images, was fine-tuned using simulated neutrino images and when trained with one hundred thousand training events reached an F1 score of $0.896 \pm 0.002$ compared to $0.836 \pm 0.004$ from a randomly-initialised network trained with the same training sample. The transfer-learned networks also demonstrate lower bias as a function of energy and more balanced performance across different interaction types.