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 2603.06008cond-mat.stat-mech
收藏

通过深度学习方法改进的动态标度方法

Dynamical scaling method improved by a deep learning approach

The University of Electro-Communications(电气通信大学)

Yusuke Terasawa, Yukiyasu Ozeki

AI总结:

本文提出一种改进的动态标度分析方法,利用深度学习降低计算成本,应用于二维伊辛模型和三维状态Potts模型,实现了更高的精度和效率。

摘要:

我们提出了一种通过深度学习方法改进的动态标度分析。尽管高斯过程回归已被广泛用于估计标度参数,但在涉及大数据集的动态标度分析中,其参数优化的计算成本成为限制因素。相比之下,本文的方法采用神经网络,显著降低了计算成本,并能够利用传统高斯过程回归无法访问的完整数据集。我们将其应用于二维伊辛模型和二维三维状态Potts模型,实现了比传统方法更高的精度和计算效率。

英文摘要:

We propose a dynamical scaling analysis improved by a deep learning approach. While Gaussian process regression has been widely employed for estimating scaling parameters, its computational cost for parameter optimization becomes a limitation in dynamical scaling analysis, where large datasets are involved. In contrast, the present method employs a neural network, which significantly reduces the computational cost and enables the use of the entire dataset that was inaccessible with Gaussian process regression. We applied the method to the 2D Ising model and the 2D 3-state Potts model, achieving higher accuracy and computational efficiency than conventional approaches.

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。

↑