Uncertainty quantification of spectral predictions using deep neural networks.
Uncertainty quantification of spectral predictions using deep neural networks.
复制标题
使用深度神经网络对光谱预测的不确定性进行量化。
DOI：
10.1039/d3cc01988h
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
Verma S
中科院分区：
文献类型：
--
作者：
Verma S
研究了深度集成和Bootstrap重采样两种不确定量化方法对过渡金属K边X射线吸收近边结构(XANES)谱的深度神经网络(DNN)预测的性能。Bootstrap重采样与我们的多层感知器模型相结合，提供了准确的不确定度评估，所有预测光谱强度的90%落在9个第一行过渡金属K边XANES光谱保持数据真实值的±3σ以内。
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 ±3σ of the true values for held-out data across the nine first-row transition metal K-edge XANES spectra.