Transfer Learning for Neutrino Scattering: Domain Adaptation with GANs
This provides a domain-specific solution for neutrino physics researchers to improve event generators in data-scarce scenarios, but it is incremental as it builds on existing GAN and transfer learning methods.
The paper tackled the problem of generating neutrino scattering events with sparse experimental data by using transfer learning to adapt a GAN model from synthetic charged-current neutrino-carbon data to other interactions like neutrino-argon and antineutrino-carbon, resulting in models that significantly outperformed training from scratch, even with as few as 10,000 events.
We utilize transfer learning to extrapolate the physics knowledge encoded in a Generative Adversarial Network (GAN) model trained on synthetic charged-current (CC) neutrino-carbon inclusive scattering data. This base model is adapted to generate CC inclusive scattering events (lepton kinematics only) for neutrino-argon and antineutrino-carbon interactions. Furthermore, we assess the effectiveness of transfer learning in re-optimizing a custom model when new data comes from a different neutrino-nucleus interaction model. Our results demonstrate that transfer learning significantly outperforms training generative models from scratch. To study this, we consider two training data sets: one with 10,000 and another with 100,000 events. The models obtained via transfer learning perform well even with smaller training data. The proposed method provides a promising approach for constructing neutrino scattering event generators in scenarios where experimental data is sparse.