Deep Learning Methods On Neutron Scattering Data

Abstract

Recently, by using deep learning methods, a computer is able to surpass or come close to matching human performance on image analysis and recognition. This advanced methods could also help extracting features from neutron scattering experimental data. Those data contain rich scientific information about structure and dynamics of materials under investigation. Deep learning could help researchers better understand the link between experimental data and materials properties. Moreover,it could also help to optimize neutron scattering experiment by predicting the best possible instrument configuration. Among all possible experimental methods, we begin our study on the small-angle neutron scattering (SANS) data and by predicting the structure geometry of the sample material at an early stage. This step is a keystone to predict the experimental parameters to properly setup the instrument as well as the best measurement strategy. In this paper, we propose to use transfer learning to retrain a convolutional neural networks (CNNs) based pre rained model to adapt the scattering images classification, which could predict the structure of the materials at an early stage in the SANS experiment. This deep neural network is trained and validated on simulated database, and tested on real scattering images.

Machine learning for analysis of experimental scattering and spectroscopy data in materials chemistry

“…Supervised ML has also proved to be an efficient tool for direct parameter extraction from SAS data (main application 3, Fig. 2), which might be difficult or time-consuming for humans to detect, such as orientation, 68 shape, [69][70][71] or the model for SAS form factor tting. [72][73][74] For example, the Scattering Ai aNalysis (SCAN) tool can predict the model for SAS form factor tting from a SAXS pattern obtained from a nanoparticle.…”

Learning from virtual experiments to assist users of Small Angle Neutron Scattering in model selection

“…While an averaged particle size is easily available from SANS data, detailed information about size distribution and geometry are difficult to obtain, making any support by Machine Learning (ML) algorithms desirable. Some efforts in using ML algorithms for SANS data have already been made in one-dimensional curves 9,10 and in two-dimensional images 11 showing great potential on small datasets. Nevertheless, algorithms like deep neural networks require large datasets and this becomes a problem in neutron science because conducting experiments in real-world settings is time-consuming (due to complex sample preparations, experimental setups, and data collection processes).…”

Machine learning for scattering data: strategies, perspectives and applications to surface scattering

“…For example, work on bulk crystallography started many years ago and has showed impressive progress (Tatlier, 2011;Oviedo et al, 2019;Lee et al, 2020;Bai et al, 2018). Other standard scattering methods, such as small-angle scattering, have also received considerable attention, especially for classification tasks (Song et al, 2020;Ikemoto et al, 2020;Franke et al, 2018;Archibald et al, 2020;Chang et al, 2020). For nonstandard coherent scattering methods, such as X-ray photon correlation spectroscopy (XPCS), ML-based analysis using autoencoders has been employed (Konstantinova et al, 2021;Timmermann et al, 2022).…”