arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2603.15853cond-mat.supr-concond-mat.dis-nnquant-ph
收藏

驯服神经网络波函数的表达性以实现对量子多体态的稳健收敛

Taming the expressiveness of neural-network wave functions for robust convergence to quantum many-body states

Pennsylvania State University(宾夕法尼亚州立大学)

Dezhe Z. Jin

AI总结:

本文提出通过最小化局部能量方差的对数压缩来提升神经网络波函数的收敛性,并展示其在二维谐振子陷阱中自旋1/2粒子的应用。

摘要:

神经网络正逐渐成为确定相互作用多体费米系统量子态的强大工具。标准方法通过最小化从蒙特卡罗样本估计的平均局部能量来训练神经网络近似方案。然而,这通常导致估计平均能量的样本间波动较大,从而减缓能量最小化的收敛速度。本文提出通过最小化局部能量方差的对数压缩可显著提升收敛性。此外,这种损失函数可被调整以系统地获得多次运行中的能量谱。我们展示了这些想法在二维谐振子陷阱中自旋1/2粒子的吸引性Poschl-Teller相互作用下的应用。

英文摘要:

Neural networks are emerging as a powerful tool for determining the quantum states of interacting many-body fermionic systems. The standard approach trains a neural-network ansatz by minimizing the mean local energy estimated from Monte Carlo samples. However, this typically results in large sample-to-sample fluctuations in the estimated mean energy and thus slow convergence of the energy minimization. We propose that minimizing a logarithmically compressed variance of the local energies can dramatically improve convergence. Moreover, this loss function can be adapted to systematically obtain the energy spectrum across multiple runs. We demonstrate these ideas for spin-1/2 particles in a 2D harmonic trap with attractive Poschl-Teller interactions between opposite spins.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑