Transfer Learning for Neutrino Scattering: Domain Adaptation with GANs

Abstract

Transfer learning (TL) is used to extrapolate the physics information encoded in a Generative Adversarial Network (GAN) trained on synthetic neutrino-carbon inclusive scattering data to related processes such as neutrino-argon and antineutrino-carbon interactions. We investigate how much of the underlying lepton-nucleus dynamics is shared across different targets and processes. We also assess the effectiveness of TL when training data is obtained from a different neutrino-nucleus interaction model. Our results show that TL not only reproduces key features of lepton kinematics, including the quasielastic and $\Delta$-resonance peaks, but also significantly outperforms generative models trained from scratch. Using data sets of 10,000 and 100,000 events, we find that TL maintains high accuracy even with limited statistics. Our findings demonstrate that TL provides a well-motivated and efficient framework for modeling (anti)neutrino-nucleus interactions and for constructing next-generation neutrino-scattering event generators, particularly valuable when experimental data are sparse.

23 pages, 22 figures, together with supplement, as published in Phys. Rev. D

References in corpus (21)

• The GENIE Neutrino Monte Carlo Generator
• NuSTEC White Paper: Status and Challenges of Neutrino-Nucleus Scattering
• Learning Particle Physics by Example: Location-Aware Generative Adversarial Networks for Physics Synthesis
• Physics Potential of a Long Baseline Neutrino Oscillation Experiment Using J-PARC Neutrino Beam and Hyper-Kamiokande
• Final State Interactions Effects in Neutrino-Nucleus Interactions
• The NEUT Neutrino Interaction Simulation
• Machine Learning and LHC Event Generation
• $C_5^A$ axial form factor from bubble chamber experiments
• Neutrino-Nucleon Cross-Section Model Tuning in GENIE v3
• Simulations from a new neutrino event generator
• Event Generators for High-Energy Physics Experiments
• Deep Learning as a Parton Shower
• ACHILLES: A novel event generator for electron- and neutrino-nucleus scattering
• Determination of the argon spectral function from (e,e'p) data
• Towards a Deep Learning Model for Hadronization
• Determination of the titanium spectral function from (e,e'p) data
• Cross Sections for A(e,e')X Reactions
• WroNG -- Wroclaw Neutrino Generator of events for single pion production
• Application of Transfer Learning to Neutrino Interaction Classification
• JLab spectral functions of argon in NuWro and their implications for MicroBooNE
• Electron-nucleus cross sections from transfer learning

Cited by in corpus (1)

• Bayesian Reasoning for Physics Informed Neural Networks