Enhancing Accuracy of Deep Learning Algorithms by Training with Low-Discrepancy Sequences


METADATA ONLY

Date

2021

Publication Type

Journal Article

ETH Bibliography

yes

Citations

Altmetric
METADATA ONLY

Data

Rights / License

Abstract

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. © 2021, Society for Industrial and Applied Mathematics

Permanent link

Publication status

published

Editor

Book title

Volume

59 (3)

Pages / Article No.

1811 - 1834

Publisher

SIAM

Event

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

deep learning; deep neural networks; low-discrepancy sequences; Quasi-Monte Carlo

Organisational unit

03851 - Mishra, Siddhartha / Mishra, Siddhartha check_circle

Notes

Funding

770880 - Computation and analysis of statistical solutions of fluid flow (EC)

Related publications and datasets

Is variant form of: