Iterative peak-fitting of frequency-domain data via deep convolution neural networks

TL;DR

This paper develops a deep convolution neural network approach, specifically a novel SENet variant, for automated peak fitting in frequency-domain spectroscopic data, improving analysis speed and accuracy for material discovery.

Contribution

Introduces a new SENet-based deep learning model for iterative peak fitting in one-dimensional spectral data, outperforming other architectures and enhancing automated analysis capabilities.

Findings

SENet variant achieved the best performance among tested architectures.

Deep learning model effectively decomposed noisy spectral data into multiple peaks.

Application to experimental spectra demonstrated practical utility and limitations.

Abstract

High-throughput material screening for the discovery and design of novel functional materials requires automatized analyses of theoretical and experimental data. Here we study the subject of human-free analyses of one-dimensional spectroscopic data, {\it e.g.} in the frequency domain, via employing deep convolution neural network. Specifically, we trained various deep convolution neural network and benchmarked their performance in decomposing one-dimensional noisy data into multiple nonorthogonal peaks in an iterative manner, after which a conventional basin-hopping algorithm was applied to further reduce residual fitting error. Among six different network architectures, a variant of "Squeeze-and-excitation" network (SENet) structure that we first propose in this study shows the best performance. Dependency of training performance with respect to the choice of the loss function is also discussed. We conclude by applying our modified SENet model to experimental photoemission spectra of graphene, MoS$_2$, and WS$_2$ and address its potential applications and limitations.