Machine learning-based inversion of nuclear responses

Abstract

A microscopic description of the interaction of atomic nuclei with external electroweak probes is required for elucidating aspects of short-range nuclear dynamics and for the correct interpretation of neutrino oscillation experiments. Nuclear quantum Monte Carlo methods infer the nuclear electroweak response functions from their Laplace transforms. Inverting the Laplace transform is a notoriously ill-posed problem; and Bayesian techniques, such as maximum entropy, are typically used to reconstruct the original response functions in the quasielastic region. In this work, we present a physics-informed artificial neural network architecture suitable for approximating the inverse of the Laplace transform. Utilizing simulated, albeit realistic, electromagnetic response functions, we show that this physics-informed artificial neural network outperforms maximum entropy in both the low-energy transfer and the quasielastic regions, thereby allowing for robust calculations of electron scattering and neutrino scattering on nuclei and inclusive muon capture rates.

arXiv comment12 pages, 10 figures

References in corpus (42)

• Adam: A Method for Stochastic Optimization
• Machine learning and the physical sciences
• Quantum Monte Carlo methods for nuclear physics
• Observation of Coherent Elastic Neutrino-Nucleus Scattering
• A Review of Convolutional Neural Networks for Inverse Problems in Imaging
• Nucleon-Nucleon Correlations, Short-lived Excitations, and the Quarks Within
• NuSTEC White Paper: Status and Challenges of Neutrino-Nucleus Scattering
• Nuclear Mass Predictions for the Crustal Composition of Neutron Stars: A Bayesian Neural Network Approach
• Inclusive quasi-elastic electron-nucleus scattering
• Response Functions from Integral Transforms with a Lorentz Kernel
• Electric dipole polarizability of ${}^{48}$Ca and implications for the neutron skin
• Nucleon and nucleon-pair momentum distributions in A <= 12 nuclei
• Neutron drip line in the Ca region from Bayesian model averaging
• Superscaling of Inclusive Electron Scattering from Nuclei
• Nuclear charge radii: Density functional theory meets Bayesian neural networks
• Longitudinal and Transverse Quasi-Elastic Response Functions of Light Nuclei
• Elastic Electron Scattering from Light Nuclei
• Electromagnetic response of 12C: a first-principles calculation
• Artificial Neural Network Approach to the Analytic Continuation Problem
• Variational Monte Carlo calculations of $\mathbf{A}\le 4$ nuclei with an artificial neural-network correlator ansatz
• First principles description of the giant dipole resonance in 16O
• Inclusive neutrino scattering off deuteron from threshold to GeV energies
• The detection of back-to-back proton pairs in Charged-Current neutrino interactions with the ArgoNeuT detector in the NuMI low energy beam line
• Analytic continuation via domain-knowledge free machine learning
• Total Photoabsorption Cross Sections of A=6 Nuclei with Complete Final State Interaction
• Decoding Beta-Decay Systematics: A Global Statistical Model for Beta^- Halflives
• Neutrino-nucleus interactions and the determination of oscillation parameters
• Ab initio calculation of neutral-current $\nu$-${}^{12}$C inclusive quasielastic scattering
• Deep learning: Extrapolation tool for ab initio nuclear theory
• Coherent elastic neutrino-nucleus scattering on 40Ar from first principles
• Machine learning the deuteron
• Extrapolation of nuclear structure observables with artificial neural networks
• Ab initio study of $({\mathbf{\nu }}_{\mathbf{\ell }},{\mathbf{\ell }}^{-})$ and $({\overline{\mathbf{\nu }}}_{\mathbf{\ell }},{\mathbf{\ell }}^{+})$ inclusive scattering in ${}^{12}$C: confronting the MiniBooNE and T2K CCQE data
• Electron- versus neutrino-nucleus scattering
• Photoemission spectra of many-polaron systems
• Role of final state interaction and of three-body force on the longitudinal response function of 4He
• Spectral density estimation with the Gaussian Integral Transform
• Accurate Four-Body Response Function with Full Final State Interaction: Application to Electron Scattering off 4He
• Muon capture in nuclei: an ab initio approach based on quantum Monte Carlo methods
• Momentum space treatment of inclusive neutrino scattering off the deuteron and trinucleons
• Electromagnetic scaling functions within the Green's Function Monte Carlo approach
• Analytic Continuation of Noisy Data Using Adams Bashforth ResNet

Cited by in corpus (17)

• Machine Learning in Nuclear Physics
• Physically Interpretable Machine Learning for nuclear masses
• Nuclear energy density functionals from machine learning
• Electron scattering on $\mathbf{A}=3$ nuclei from quantum Monte Carlo based approaches
• Spectral density reconstruction with Chebyshev polynomials
• Machine learning approach to pattern recognition in nuclear dynamics from the ab initio symmetry-adapted no-core shell model
• Nuclear Structure at the Crossroads
• Theoretical tools for neutrino scattering: interplay between lattice QCD, EFTs, nuclear physics, phenomenology, and neutrino event generators
• Demonstration of the Rodeo Algorithm on a Quantum Computer
• Machine-learning approach to finite-size effects in systems with strongly interacting fermions
• Faster spectral density calculation using energy moments
• Neural Network predictions of inclusive electron-nucleus cross sections
• Spectral function for ${}^4$He using the Chebyshev expansion in coupled-cluster theory
• Machine learning the deuteron: new architectures and uncertainty quantification
• Controlled Gate Networks: Theory and Application to Eigenvalue Estimation
• Nuclear two point correlation functions on a quantum-computer
• A Parameter-Free Differential Evolution Algorithm for the Analytic Continuation of Imaginary Time Correlation Functions