DS4MS
Data Science for Multiplexing Spectrometers
Abstract
The multiplexing neutron spectrometers at PSI enables to collect high-dimensional neutron scattering data. Existing methods do not utilize such datasets to their full potential, are time consuming, and require significant expert input. The project’s goal is to develop a generic, automatized method of fitting a signal simulation directly to a high-dimensional dataset, even when the simulation is computationally expensive, and even in the presence of background for which no analytic model exists. This data analysis tool will furthermore improve the data collection process during experiments through an automated intelligent decision making algorithm that will determine when a sufficient amount of neutrons has been collected in an arbitrary measurement setting. This work will enable research on new specific phenomena, free up time for instrument scientists, and ensure the optimal use of limited beam time.
People
Collaborators
Victor has joined the SDSC in 2020 to design solutions for data-driven optimization problems. His research interests lie at the crossroad of machine learning and decision-making. This contains several topics such as stochastic optimization, reinforcement learning, combinatorial optimization, and probabilistic graphical models. Victor received a PhD in operations research and machine learning from Ecole des Ponts Paristech in 2020. Before that, he completed a master degree in Operation Research and Machine learning at Ecole des Ponts Paristech and a bachelor degree in Applied Mathematics and Computer Sciences.
Benjamín Béjar received a PhD in Electrical Engineering from Universidad Politécnica de Madrid in 2012. He served as a postdoctoral fellow at École Polytechnique Fédérale de Lausanne until 2017, and then he moved to Johns Hopkins University where he held a Research Faculty position until Dec. 2019. His research interests lie at the intersection of signal processing and machine learning methods, and he has worked on topics such as sparse signal recovery, time-series analysis, and computer vision methods with special emphasis on biomedical applications. Since 2021, Benjamin leads the SDSC office at the Paul Scherrer Institute in Villigen.
PI | Partners:
description
Motivation
The goal of the project is to develop a generic, automized method of fitting a signal of neutron scattering data in the presence of background noise. The first goal is to capture separately the signal and the background noise. We then aim at improving the data collection process through an intelligent decision making algorithm that will determine when a sufficient amount of neutrons has been collected in an arbitrary measurement setting.
Proposed Approach / Solution
SDSC develops ML models and optimization algorithms to tackle the fitting problem and the decision-making problem. First, the approach provides a denoising algorithm that extracts the signal from the noisy observations (Fig. 2). The solution is based on the resolution of a regularized problem that leverages the rotation invariance property of the background noise. Second, the approach identifies the regions that require more measurement to get a good signal-to-noise ratio. We create a stoping criterion ensuring that enough have been collected to get a good signal-to-noise ratio. The solution is implemented in the MJOLNIR software that treats the collected data.
Impact
The outcomes of the project will help to exploit collected data and further improve the use of the multiplexing instrument (Fig. 1) by reducing beam time and by providing more insights in the scattering data analysis. This approach could be also adapted for different application including time-of-light neutron spectroscopy.
Presentation
Gallery
Annexe
Additional resources
Bibliography
- Lass, J., et al. Design and performance of the multiplexing spectrometer CAMEA.
- arXiv:2007.14796 [physics.ins-det]
- Allenspach, S., et al. (2021). Revealing three-dimensional quantum criticality by Sr
- substitution in Han purple. Physical Review Research, 3, 023177.
- Lass, J., Jacobsen, H., Mazzone, D. G., & Lefmann, K. (2020). MJOLNIR: a software package for multiplexing neutron spectrometers. SoftwareX, 12, 100600.
- Noack, M. et al. (2021). Gaussian processes for autonomous data acquisition at large-scale synchrotron and neutron facilities. Nature Reviews Physics, 3, 685.
Publications
Related Pages
More projects
ML-L3DNDT
BioDetect
IRMA
FLBI
News
Latest news
Smartair | An active learning algorithm for real-time acquisition and regression of flow field data
Smartair | An active learning algorithm for real-time acquisition and regression of flow field data
The Promise of AI in Pharmaceutical Manufacturing
The Promise of AI in Pharmaceutical Manufacturing
Efficient and scalable graph generation through iterative local expansion
Efficient and scalable graph generation through iterative local expansion
Contact us
Let’s talk Data Science
Do you need our services or expertise?
Contact us for your next Data Science project!