A Boson-informed neural network for Ground States of Many-Body Schrödinger equations
Abstract
We propose a symmetry-preserving Boson-informed neural network (BINN) for the variational Monte Carlo computation of interacting-boson ground states in continuous space. The framework explicitly separates the two dominant physical scales of the system. At the microscopic interaction scale, a cell Jastrow factor absorbs the leading near-contact stiffness, while at the macroscopic scale, a potential-aware envelope encodes the external confinement, leaving the neural network to capture a smoother many-body residual. Across both two- and three-dimensional tests, the BINN approaches the Hartree and Gross-Pitaevskii limits with an empirical 𝒪(N−1) finite-size trend in both energy and density. The energies agree with diffusion Monte Carlo to within 0.06% for the particle number tested up to $350$ in the three-dimensional shrinking-range case. Applications are made under various trapping scenarios, justifying robust performance of BINN.
Related
- Solving Fredholm Integral Equations with Physics-Informed Neural Networks: Single-and double-network approaches
- CI-PINN: Causal Integral Physics-Informed Neural Network for Solving Evolution Equations
- QGPINNs: A Physics-Informed Neural Network Framework for Nonlocal Differential Equations on Quantum Graphs
- Ground states for pseudo-relativistic Schrödinger equations with logarithm-law: qualitative properties and nonrelativistic limit
- Sharp stability near sums of ground states for fractional Schrödinger equations
BibTeX
@misc{kong2026bosoninformed,
title = {{A Boson-informed neural network for Ground States of Many-Body Schrödinger equations}},
author = {Zhizhong Kong and Jerry Zhijian Yang and Cheng Yuan and Xiaofei Zhao},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.18875},
eprint = {2610.18875},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.18875},
}
RIS
TY - UNPB
TI - A Boson-informed neural network for Ground States of Many-Body Schrödinger equations
AU - Zhizhong Kong
AU - Jerry Zhijian Yang
AU - Cheng Yuan
AU - Xiaofei Zhao
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.18875
UR - https://arxiv.org/abs/2610.18875
AB - We propose a symmetry-preserving Boson-informed neural network (BINN) for the variational Monte Carlo computation of interacting-boson ground states in continuous space. The framework explicitly separates the two dominant physical scales of the system. At the microscopic interaction scale, a cell Jastrow factor absorbs the leading near-contact stiffness, while at the macroscopic scale, a potential-aware envelope encodes the external confinement, leaving the neural network to capture a smoother many-body residual. Across both two- and three-dimensional tests, the BINN approaches the Hartree and Gross-Pitaevskii limits with an empirical 𝒪(N^−1) finite-size trend in both energy and density. The energies agree with diffusion Monte Carlo to within 0.06% for the particle number tested up to $350$ in the three-dimensional shrinking-range case. Applications are made under various trapping scenarios, justifying robust performance of BINN.
ER -