English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

High-throughput material screening for the discovery and design of novelfunctional materials requires automatized analyses of theoretical andexperimental data. Here we study the subject of human-free analyses ofone-dimensional spectroscopic data, {\it e.g.} in the frequency domain, viaemploying deep convolution neural network. Specifically, we trained variousdeep convolution neural network and benchmarked their performance indecomposing one-dimensional noisy data into multiple nonorthogonal peaks in aniterative manner, after which a conventional basin-hopping algorithm wasapplied to further reduce residual fitting error. Among six different networkarchitectures, a variant of "Squeeze-and-excitation" network (SENet) structurethat we first propose in this study shows the best performance. Dependency oftraining performance with respect to the choice of the loss function is alsodiscussed. We conclude by applying our modified SENet model to experimentalphotoemission spectra of graphene, MoS$_2$, and WS$_2$ and address itspotential applications and limitations.

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