SIAM Journal on Numerical Analysis, Volume 59, Issue 3, Page 1811-1834, January 2021.
We propose a supervised deep learning algorithm based on low-discrepancy sequences as the training set. By a combination of theoretical arguments and extensive numerical experiments we demonstrate that the proposed algorithm significantly outperforms standard deep learning algorithms that are based on randomly chosen training data for problems in moderately high dimensions. The proposed algorithm provides an efficient method for building inexpensive surrogates for many underlying maps in the context of scientific computing.

中文翻译：

---

通过低差异序列训练提高深度学习算法的准确性

SIAM Journal on Numerical Analysis，第 59 卷，第 3 期，第 1811-1834 页，2021 年 1 月。
我们提出了一种基于低差异序列作为训练集的监督深度学习算法。通过理论论证和大量数值实验的结合，我们证明所提出的算法显着优于基于随机选择训练数据的标准深度学习算法，以解决中等高维问题。所提出的算法为在科学计算的背景下为许多底层地图构建廉价代理提供了一种有效的方法。