Uncertainty quantification of spectral predictions using deep neural networks

被引：4

作者：

Verma, Sneha ^{[1
]}

Aznan, Nik Khadijah Nik ^{[2
]}

Garside, Kathryn ^{[2
]}

Penfold, Thomas J. ^{[1
]}

机构：

[1] Newcastle Univ, Chem, Sch Nat & Environm Sci, Newcastle Upon Tyne NE1 7RU, England

[2] Newcastle Univ, Res Software Engineer Grp, Newcastle Upon Tyne NE1 7RU, England

来源：

CHEMICAL COMMUNICATIONS | 2023年 / 59卷 / 46期

基金：

英国工程与自然科学研究理事会;

关键词：

LIGHT;

D O I：

10.1039/d3cc01988h

中图分类号：

O6 [化学];

学科分类号：

0703 ;

摘要：

We investigate the performance of uncertainty quantification methods, namely deep ensembles and bootstrap resampling, for deep neural network (DNN) predictions of transition metal K-edge X-ray absorption near-edge structure (XANES) spectra. Bootstrap resampling combined with our multi-layer perceptron (MLP) model provides an accurate assessment of uncertainty with >90% of all predicted spectral intensities falling within +/- 3 sigma of the true values for held-out data across the nine first-row transition metal K-edge XANES spectra.

引用

页码：7100 / 7103

页数：4

共 50 条

[1] Uncertainty quantification for predictions of atomistic neural networks
Vazquez-Salazar, Luis Itza
Boittier, Eric D.
Meuwly, Markus
[J]. CHEMICAL SCIENCE, 2022, 13 (44) : 13068 - 13084
[2] Predictive Uncertainty Quantification of Deep Neural Networks using Dirichlet Distributions
Hammam, Ahmed
Bonarens, Frank
Ghobadi, Seyed Eghbal
Stiller, Christoph
[J]. PROCEEDINGS OF 6TH ACM COMPUTER SCIENCE IN CARS SYMPOSIUM, CSCS 2022, 2022,
[3] Measuring the Uncertainty of Predictions in Deep Neural Networks with Variational Inference
Steinbrener, Jan
Posch, Konstantin
Pilz, Juergen
[J]. SENSORS, 2020, 20 (21) : 1 - 22
[4] Leveraging the Bhattacharyya coefficient for uncertainty quantification in deep neural networks
Van Molle, Pieter
Verbelen, Tim
Vankeirsbilck, Bert
De Vylder, Jonas
Diricx, Bart
Kimpe, Tom
Simoens, Pieter
Dhoedt, Bart
[J]. NEURAL COMPUTING & APPLICATIONS, 2021, 33 (16): : 10259 - 10275
[5] Leveraging the Bhattacharyya coefficient for uncertainty quantification in deep neural networks
Pieter Van Molle
Tim Verbelen
Bert Vankeirsbilck
Jonas De Vylder
Bart Diricx
Tom Kimpe
Pieter Simoens
Bart Dhoedt
[J]. Neural Computing and Applications, 2021, 33 : 10259 - 10275
[6] Scaling Up Bayesian Uncertainty Quantification for Inverse Problems Using Deep Neural Networks
Lan, Shiwei
Li, Shuyi
Shahbaba, Babak
[J]. SIAM-ASA JOURNAL ON UNCERTAINTY QUANTIFICATION, 2022, 10 (04): : 1684 - 1713
[7] Clarifying trust of materials property predictions using neural networks with distribution-specific uncertainty quantification
Gruich, Cameron J.
Madhavan, Varun
Wang, Yixin
Goldsmith, Bryan R.
[J]. MACHINE LEARNING-SCIENCE AND TECHNOLOGY, 2023, 4 (02):
[8] Neural SDE-Based Epistemic Uncertainty Quantification in Deep Neural Networks
Tharzeen, Aabila
Dahale, Shweta
Natarajan, Balasubramaniam
[J]. ENGINEERING APPLICATIONS OF NEURAL NETWORKS, EANN 2024, 2024, 2141 : 247 - 258
[9] Large-Scale Spectral Analysis for Element Quantification using Deep Neural Networks
Xie, Xiang
Stork, Wilhelm
[J]. 2022 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2022,
[10] Bond order predictions using deep neural networks
Magedov, Sergey
Koh, Christopher
Malone, Walter
Lubbers, Nicholas
Nebgen, Benjamin
[J]. JOURNAL OF APPLIED PHYSICS, 2021, 129 (06)

← 1 2 3 4 5 →