NuCLR: Nuclear Co-Learned Representations

Niklas Nolte, Ouail Kitouni, Sokratis Trifinopoulos, Subhash Kantamneni, Mike Williams

Keywords: nuclear physics, representation learning

TL;DR: This paper presents an architecture to model multiple aspects of known nuclei to high precision, and finds hints of known nuclear physics in representation space.

Abstract: We introduce Nuclear Co-Learned Representations (NuCLR), a deep learning model that predicts various nuclear observables, including binding and decay energies, and nuclear charge radii. The model is trained using a multi-task approach with shared representations and obtains state-of-the-art performance, achieving levels of precision that are crucial for understanding fundamental phenomena in nuclear (astro)physics. We also report an intriguing finding that the learned representation of NuCLR exhibits the prominent emergence of crucial aspects of the nuclear shell model, namely the shell structure, including the well-known magic numbers, and the Pauli Exclusion Principle. This suggests that the model is capable of capturing the underlying physical principles, and that our approach has the potential to offer valuable insights into nuclear theory.

Submission Number: 75