Architecture as physical prior: cooperative neural network for nuclear masses

TL;DR

This paper introduces the Cooperative Neural Network (CoNN), a novel architecture that predicts nuclear masses directly from proton and neutron numbers by embedding physical principles into its structure, achieving high accuracy without prior feature engineering.

Contribution

The paper presents the CoNN architecture that incorporates structural inductive biases to predict nuclear masses from raw data, eliminating the need for theoretical baselines or hand-crafted features.

Findings

Achieves 0.269 MeV RMS deviation on AME2020 dataset

Develops embeddings with extrema at magic numbers

Reproduces odd--even staggering without supervision

Abstract

Machine learning approaches to nuclear mass prediction have achieved remarkable accuracy, but typically rely on existing theoretical baselines or hand-crafted physics features. Here we demonstrate that these prerequisites can be supplanted by structural inductive biases embedded directly in the network architecture. We present the Cooperative Neural Network (CoNN), which predicts binding energies from raw proton and neutron numbers (Z,N) alone by additively combining four structurally constrained modules: a smooth network for bulk liquid-drop trends, discrete scalar embeddings for shell effects, a learnable two-dimensional grid for regional collective correlations, and a parity-aware network for odd--even staggering. On the AME2020 dataset, the CoNN achieves a root-mean-square deviation of 0.269 MeV across all 3558 nuclei, with 0.419 MeV on a held-out interpolation subset and 0.728 MeV on 122 nuclei newly measured since AME2016, placing it among the most accurate baseline-free approaches to direct mass prediction. Notably, the learned embeddings develop pronounced extrema at canonical magic numbers and the pairing module reproduces the expected odd--even staggering along isotopic chains, both emerging from the data without explicit supervision. These results demonstrate that physically motivated architectural constraints can effectively substitute for feature engineering, establishing architecture as physical prior as a promising paradigm for neural-network mass modeling.