Deep learning-based statistical noise reduction for multidimensional spectral data

Abstract

In spectroscopic experiments, data acquisition in multi-dimensional phase space may require long acquisition time, owing to the large phase space volume to be covered. In such case, the limited time available for data acquisition can be a serious constraint for experiments in which multidimensional spectral data are acquired. Here, taking angle-resolved photoemission spectroscopy (ARPES) as an example, we demonstrate a denoising method that utilizes deep learning as an intelligent way to overcome the constraint. With readily available ARPES data and random generation of training data set, we successfully trained the denoising neural network without overfitting. The denoising neural network can remove the noise in the data while preserving its intrinsic information. We show that the denoising neural network allows us to perform similar level of second-derivative and line shape analysis on data taken with two orders of magnitude less acquisition time. The importance of our method lies in its applicability to any multidimensional spectral data that are susceptible to statistical noise.

arXiv comment8 pages, 8 figures

References in corpus (21)

• Adam: A Method for Stochastic Optimization
• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• Angle-resolved photoemission spectroscopy of the cuprate superconductors
• Electronic structure of quantum materials studied by angle-resolved photoemission spectroscopy
• Probing the Low-Energy Electronic Structure of Complex Systems by ARPES
• A precise method for visualizing dispersive features in image plots
• Observation of two distinct $d_{xz}/d_{yz}$ band splittings in FeSe
• Self-consistent self-energy analysis of photoemission data
• Bare electron dispersion from photoemission experiments
• New Developments in Laser-Based Photoemission Spectroscopy and its Scientific Applications: a Key Issues Review
• Neural Network-based Classification of Crystal Symmetries from X-Ray Diffraction Patterns
• Time-Resolved Momentum Microscopy with a 1 MHz High-Harmonic Extreme Ultraviolet Beamline
• One-Component Order Parameter in URu${}_2$Si${}_2$ Uncovered by Resonant Ultrasound Spectroscopy and Machine Learning
• Lifted electron pocket and reversed orbital occupancy imbalance in FeSe
• CRYSPNet: Crystal Structure Predictions via Neural Network
• Visualizing dispersive features in 2D image via minimum gradient method
• Doping-dependent energy scale of the low-energy band renormalization in (Bi,Pb)2(Sr,La)2CuO6+d
• Effects, Determination, and Correction of Count Rate Nonlinearity in Multi-Channel Analog Electron Detectors
• High repetition pump-and-probe photoemission spectroscopy based on a compact fiber laser system
• Through a Lattice Darkly -- Shedding Light on Electron-Phonon Coupling in the High T${}_c$ Cuprates
• Hidden self-energies as origin of cuprate superconductivity revealed by machine learning

Cited by in corpus (14)

• Observation of spin-dependent dual ferromagnetism in perovskite ruthenates
• A machine learning route between band mapping and band structure
• Advancing time- and angle-resolved photoemission spectroscopy: The role of ultrafast laser development
• Weak-signal extraction enabled by deep-neural-network denoising of diffraction data
• Deep learning-based denoising for fast time-resolved flame emission spectroscopy in high-pressure combustion environment
• Removing grid structure in angle-resolved photoemission spectra via deep learning method
• Transfer Learning Application of Self-supervised Learning in ARPES
• Denoising and feature extraction in photoemission spectra with variational auto-encoder neural networks
• Spectroscopic data de-noising via training-set-free deep learning method
• About two-dimensional fits for the analysis of the scattering rates and renormalization functions from angle-resolved photoelectron spectroscopy data
• An autoencoder for compressing angle-resolved photoemission spectroscopy data
• PGNAA Spectral Classification of Metal with Density Estimations
• Neural networks for a quick access to a digital twin of scanning physical properties measurements
• Machine Learning in Application to Automatic Noise Processing of Solar Spectrograms