Deep Learning strategies for ProtoDUNE raw data denoising

TL;DR

This paper explores deep learning methods, specifically graph neural networks, for denoising raw data in the ProtoDUNE experiment, aiming to improve reconstruction accuracy and computational efficiency.

Contribution

It introduces novel graph neural network architectures for denoising ProtoDUNE data and evaluates their performance against traditional algorithms.

Findings

Graph neural networks outperform traditional algorithms in denoising accuracy.

Hardware accelerators significantly speed up training and inference.

Proposed methods enhance the initial data reconstruction step.

Abstract

In this work, we investigate different machine learning-based strategies for denoising raw simulation data from the ProtoDUNE experiment. The ProtoDUNE detector is hosted by CERN and it aims to test and calibrate the technologies for DUNE, a forthcoming experiment in neutrino physics. The reconstruction workchain consists of converting digital detector signals into physical high-level quantities. We address the first step in reconstruction, namely raw data denoising, leveraging deep learning algorithms. We design two architectures based on graph neural networks, aiming to enhance the receptive field of basic convolutional neural networks. We benchmark this approach against traditional algorithms implemented by the DUNE collaboration. We test the capabilities of graph neural network hardware accelerator setups to speed up training and inference processes.