Active learning-assisted neutron spectroscopy with log-Gaussian processes

Abstract

Neutron scattering experiments at three-axes spectrometers (TAS) investigate magnetic and lattice excitations by measuring intensity distributions to understand the origins of materials properties. The high demand and limited availability of beam time for TAS experiments however raise the natural question whether we can improve their efficiency and make better use of the experimenter's time. In fact, there are a number of scientific problems that require searching for signals, which may be time consuming and inefficient if done manually due to measurements in uninformative regions. Here, we describe a probabilistic active learning approach that not only runs autonomously, i.e., without human interference, but can also directly provide locations for informative measurements in a mathematically sound and methodologically robust way by exploiting log-Gaussian processes. Ultimately, the resulting benefits can be demonstrated on a real TAS experiment and a benchmark including numerous different excitations.

arXiv commentMain: 24 pages, 6 figures, 1 table | Supplementary Information: 19 pages, 12 figures, 1 table

References in corpus (15)

• Explainable Machine Learning for Scientific Insights and Discoveries
• On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning
• Spatial and Spatio-Temporal Log-Gaussian Cox Processes: Extending the Geostatistical Paradigm
• Spin-glass ground state in a triangular-lattice compound YbZnGaO${}_4$
• Phonon self-energy and origin of anomalous neutron scattering spectra in SnTe and PbTe thermoelectrics
• Autonomous synthesis of metastable materials
• Rearrangement of uncorrelated valence bonds evidenced by low-energy spin excitations in YbMgGaO4
• Topological magnon band structure of emergent Landau levels in a skyrmion lattice
• On-the-fly Autonomous Control of Neutron Diffraction via Physics-Informed Bayesian Active Learning
• Role of vibrational entropy in the stabilization of the high-temperature phases of iron
• Robust Upward Dispersion of the Neutron Spin Resonance in the Heavy Fermion Superconductor Ce${}_{1-x}$Yb${}_x$CoIn${}_5$
• Approximate Inference for Nonstationary Heteroscedastic Gaussian process Regression
• Determining the maximum information gain and optimising experimental design in neutron reflectometry using the Fisher information
• Accelerating small-angle scattering experiments with simulation-based machine learning
• Optimising experimental design in neutron reflectometry

Cited by in corpus (5)

• Autonomous microARPES
• Resource-aware Research on Universe and Matter: Call-to-Action in Digital Transformation
• Sustaining model performance for covid-19 detection from dynamic audio data: Development and evaluation of a comprehensive drift-adaptive framework
• An Active Learning-Based Streaming Pipeline for Reduced Data Training of Structure Finding Models in Neutron Diffractometry
• Motion Planning for Triple-Axis Spectrometers