AI-Assisted GPU optimization of the Stochastic Variational Method for Few-Body Boson System
Abstract
The stochastic variational method (SVM) is one of the most powerful methods to solve quantum few-body systems precisely in various fields, such as nuclear and atomic physics. To the best of our knowledge, no SVM code optimized for GPUs has been reported. We developed a parallel SVM code for few-body clusters of 4He atoms and optimized it for two GPU architectures, NVIDIA GH200 and AMD Instinct MI300A, with the entire code written by an AI coding agent. On the algorithmic side, we introduce the secular-equation method with the Gu--Eisenstat prescription for solving the generalized eigenvalue problem within the SVM framework. The main part of the code tuning is to batch the many small matrices so that the GPUs are used efficiently, together with an array layout in memory optimized for coalesced addressing. For the five-body system with 4800 SVM basis states, the tuned SVM code runs 16.2 (14.7) times faster on the NVIDIA GH200 (AMD MI300A) than the same tuned code on the Intel Xeon Max CPU system, and 168 (153) times faster than the first working CPU implementation. The achieved performance brings 6-body and larger cluster systems within reach.
Related
- Surrogate-assisted multi-objective topology optimization design method for prefabricated large-span Suspen-Dome based on structural-mapping-method and improved-PGSA
- Stochastic Geometrically Exact Beams: Continuous-Time Geometric Principles and Variational Discretizations
- BAHAMAS: A Control Plane for Optimization and Execution of Variational Quantum Circuits
- Variational Quantum Homotopy Perturbation Method to Solve Nonlinear Partial Differential Equations
- Toward Fault-Tolerant Variational Optimization: QAOA under [[4,2,2]] Error Detection
BibTeX
@misc{shimizu2026aiassisted,
title = {{AI-Assisted GPU optimization of the Stochastic Variational Method for Few-Body Boson System}},
author = {Noritaka Shimizu and Shigeyoshi Aoyama and Yusuke Tsunoda and Akira Nukada},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.02745},
eprint = {2610.02745},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.02745},
}
RIS
TY - UNPB
TI - AI-Assisted GPU optimization of the Stochastic Variational Method for Few-Body Boson System
AU - Noritaka Shimizu
AU - Shigeyoshi Aoyama
AU - Yusuke Tsunoda
AU - Akira Nukada
PY - 2026
DA - 2026/10/02
PB - arXiv
DO - 10.48550/arxiv.2610.02745
UR - https://arxiv.org/abs/2610.02745
AB - The stochastic variational method (SVM) is one of the most powerful methods to solve quantum few-body systems precisely in various fields, such as nuclear and atomic physics. To the best of our knowledge, no SVM code optimized for GPUs has been reported. We developed a parallel SVM code for few-body clusters of ^4He atoms and optimized it for two GPU architectures, NVIDIA GH200 and AMD Instinct MI300A, with the entire code written by an AI coding agent. On the algorithmic side, we introduce the secular-equation method with the Gu--Eisenstat prescription for solving the generalized eigenvalue problem within the SVM framework. The main part of the code tuning is to batch the many small matrices so that the GPUs are used efficiently, together with an array layout in memory optimized for coalesced addressing. For the five-body system with 4800 SVM basis states, the tuned SVM code runs 16.2 (14.7) times faster on the NVIDIA GH200 (AMD MI300A) than the same tuned code on the Intel Xeon Max CPU system, and 168 (153) times faster than the first working CPU implementation. The achieved performance brings 6-body and larger cluster systems within reach.
ER -