Improved energy reconstruction in NOvA with regression convolutional neural networks

  title={Improved energy reconstruction in NOvA with regression convolutional neural networks},
  author={Pierre Baldi and Jianming Bian and Lars Hertel and Lingge Li},
  journal={Physical Review D},
In neutrino experiments, neutrino energy reconstruction is crucial because neutrino oscillations and differential cross-sections are functions of neutrino energy. It is also challenging due to the complexity in the detector response and kinematics of final state particles. We propose a regression Convolutional Neural Network (CNN) based method to reconstruct electron neutrino energy and electron energy in the NOvA neutrino experiment. We demonstrate that with raw detector pixel inputs, a… Expand
Convolutional neural networks for energy and vertex reconstruction in DUNE
Measuring neutrino CP violation and mass hierarchy is currently one of the biggest challenges in particle physics. The DUNE neutrino experiment is the next-generation flagship neutrino program in theExpand
Deep-Learning-Based Kinematic Reconstruction for DUNE
Two CNN-based methods are developed and will be presented, 2-D and 3-D, for the reconstruction of final state particle direction and energy, as well as neutrino energy, which show considerable improvements compared to the traditional methods for both scenarios. Expand
Fast inference using FPGAs for DUNE data reconstruction
The accelerator board used, a commercial off-the-shelf (COTS) hardware for fast deep learning (DL) inference based on an FPGA, and the experimental results obtained outperforming more traditional processing units are presented. Expand
Deep Learning based Energy Reconstruction for the CALICE AHCAL
Future linear lepton colliders offer the possibility of precision physics studies and demand detectors with an unprecedented resolution. For this purpose the CALICE collaboration is developingExpand
Context-enriched identification of particles with a convolutional network for neutrino events
Particle detectors record the interactions of subatomic particles and their passage through matter. The identification of these particles is necessary for in-depth physics analysis. While particlesExpand
Convolutional Neural Network for Centrality Determination in Fixed Target Experiments
Fixed-target experiments have a unique possibility to measure the centrality of colliding systems by hadronic calorimeters on the beam line. This is usually achieved by detection of all forwardExpand
Results and Prospects from NOvA
  • J. Bian
  • Physics
  • Proceedings of The 20th International Workshop on Neutrinos — PoS(NuFACT2018)
  • 2019
NOvA is a long-baseline neutrino experiment that uses an upgraded NuMI neutrino source at Fermilab and a 14-kton detector at Ash River, Minnesota. The detector has a highly active, finely segmentedExpand
Sparse autoregressive models for scalable generation of sparse images in particle physics
Generation of simulated data is essential for data analysis in particle physics, but current Monte Carlo methods are very computationally expensive. Deep-learning-based generative models haveExpand
Neural network-featured timing systems for radiation detectors: performance evaluation based on bound analysis
  • P. Ai, Z. Deng, Y. Wang, L. Li
  • Physics
  • Journal of Instrumentation
  • 2021
Waveform sampling systems are used pervasively in the design of front end electronics for radiation detection. The introduction of new feature extraction algorithms (eg. neural networks) to waveformExpand
Enhanced complex local frequency elastography method for tumor viscoelastic shear modulus reconstruction
Compared to conventional methods, the improved EC-LFE algorithm was more accurate and sensitive for small tumor detection and exhibited higher noise immunity. Expand


Electron Neutrino Energy Reconstruction in NOvA Using CNN Particle IDs
  • Shiqi Yu
  • Physics, Computer Science
  • ArXiv
  • 2019
The methods to identify particles with prong-CNN and the following approach to estimate $\nu_e$ energy for signal events are described and improving the resolution and systematic robustness of NOvA's energy estimator will improve the sensitivity of the oscillation parameters measurement. Expand
A Convolutional Neural Network Neutrino Event Classifier
This algorithm, CVN (Convolutional Visual Network) identifies neutrino interactions based on their topology without the need for detailed reconstruction and outperforms algorithms currently in use by the NOvA collaboration. Expand
Deep Neural Networks for Energy and Position Reconstruction in EXO-200
This work describes a unique approach that, in an experiment such as EXO-200, allows to successfully perform certain reconstruction and analysis tasks by training the network on waveforms from experimental data, either reducing or eliminating the reliance on the Monte Carlo. Expand
Convolutional neural networks applied to neutrino events in a liquid argon time projection chamber
We present several studies of convolutional neural networks applied to data coming from the MicroBooNE detector, a liquid argon time projection chamber (LArTPC). The algorithms studied include theExpand
The NOvA simulation chain
The NOνA experiment is a two-detector, long-baseline neutrino experiment operating in the recently upgraded NuMI muon neutrino beam. Simulating neutrino interactions and backgrounds requires manyExpand
Revealing Fundamental Physics from the Daya Bay Neutrino Experiment Using Deep Neural Networks
It is shown that meaningful physical content can be revealed by transforming the raw data into a learned high-level representation using deep neural networks, with measurements taken at the Daya Bay Neutrino Experiment as a case study. Expand
Observation of Electron Neutrino Appearance in the NuMI Beam with the NOvA Experiment
NOvA is a long-baseline neutrino oscillation experiment that uses two functionally identical detectors separated by 810 kilometers at locations 14 milliradians off-axis from the NuMI muon neutrinoExpand
Event Reconstruction Techniques in NOvA
The NOvA experiment is a long-baseline neutrino oscillation experiment utilizing the NuMI beam generated at Fermilab. The experiment will measure the oscillations within a muon neutrino beam in a 300Expand
Background rejection in NEXT using deep neural networks
We investigate the potential of using deep learning techniques to reject background events in searches for neutrinoless double beta decay with high pressure xenon time projection chambers capable ofExpand
Jet Substructure Classification in High-Energy Physics with Deep Neural Networks
At the extreme energies of the Large Hadron Collider, massive particles can be produced at such high velocities that their hadronic decays are collimated and the resulting jets overlap. DeducingExpand