Bayesian Learning of Adatom Interactions from Atomically Resolved Imaging Data

Accelerated Nano-Optical Imaging through Sparse Sampling

The integration time and signal-to-noise ratio are inextricably linked when performing scanning probe microscopy based on raster scanning. This often yields a large lower bound on the measurement time, for example, in nano-optical imaging experiments performed using a scanning near-field optical microscope (SNOM). Here, we utilize sparse scanning augmented with Gaussian process regression to bypass the time constraint. We apply this approach to image charge-transfer polaritons in graphene residing on ruthenium trichloride (α-RuCl3) and obtain key features such as polariton damping and dispersion. Critically, nano-optical SNOM imaging data obtained via sparse sampling are in good agreement with those extracted from traditional raster scans but require 11 times fewer sampled points. As a result, Gaussian process-aided sparse spiral scans offer a major decrease in scanning time.

Machine learning of all-dielectric core-shell nanostructures: the critical role of the objective function in inverse design

To integrate nanophotonics into light-based technologies, it is critical to elicit a desired optical response from its fundamental component, a nanoresonator. Because the optical resonance of a nanoresonator depends strongly on its base material and structural features, machine learning has been contemplated to enhance the design and optimization processes. However, its accuracy in searching the vast parameter space of nanophotonics still poses unresolved questions. Here, we show how the choice of objective functions, in combination with trained neural networks, can drastically change the optimization process-even for a simple nanophotonic structure. To assess how different objective functions select the correct structural parameters that generate a desired optical Mie response, we use a simple core-shell, all-dielectric nanostructure as the benchmark. By controlling the proportion of training data, which represents the "experience" level, we also quantify how the various objective functions perform in finding the ground-truth parameters. Our findings demonstrate that certain objective functions exhibit improved accuracy when used with highly "experienced" neural networks. Surprisingly, we also find other objective functions that perform better when paired with less "experienced" neural networks. Taken together, our results emphasize that it is critical to understand how neural networks are coupled to optimization schemes, as is evident even when a simple core-shell nanostructure is used.

Machine Learning for Optical Scanning Probe Nanoscopy

The ability to perform nanometer-scale optical imaging and spectroscopy is key to deciphering the low-energy effects in quantum materials, as well as vibrational fingerprints in planetary and extraterrestrial particles, catalytic substances, and aqueous biological samples. These tasks can be accomplished by the scattering-type scanning near-field optical microscopy (s-SNOM) technique that has recently spread to many research fields and enabled notable discoveries. Herein, it is shown that the s-SNOM, together with scanning probe research in general, can benefit in many ways from artificial-intelligence (AI) and machine-learning (ML) algorithms. Augmented with AI- and ML-enhanced data acquisition and analysis, scanning probe optical nanoscopy is poised to become more efficient, accurate, and intelligent.

Deciphering Alloy Composition in Superconducting Single-Layer FeSe1-xSx on SrTiO3(001) Substrates by Machine Learning of STM/S Data

Scanning tunneling microscopy (STM) is a powerful technique for imaging atomic structure and inferring information on local elemental composition, chemical bonding, and electronic excitations. However, a plain visual analysis of STM images can be challenging for such determination in multicomponent alloys, particularly beyond the diluted limit due to chemical disorder and electronic inhomogeneity. One viable solution is to use machine learning to analyze STM data and identify hidden patterns and correlations. Here, we apply this approach to determine the Se/S concentration in superconducting single-layer FeSe_1-xS_x alloys epitaxially grown on SrTiO₃(001) substrates via molecular beam epitaxy. First, the K-means clustering method is applied to identify defect-related dI/dV tunneling spectra taken by current imaging tunneling spectroscopy. Then, the Se/S ratio is calculated by analyzing the remaining spectra based on the singular value decomposition method. Such analysis provides an efficient and reliable determination of alloy composition and further reveals the correlations of nanoscale chemical inhomogeneity to superconductivity in single-layer iron chalcogenide films.

Understanding Heterogeneities in Quantum Materials

Quantum materials are usually heterogeneous, with structural defects, impurities, surfaces, edges, interfaces, and disorder. These heterogeneities are sometimes viewed as liabilities within conventional systems; however, their electronic and magnetic structures often define and affect the quantum phenomena such as coherence, interaction, entanglement, and topological effects in the host system. Therefore, a critical need is to understand the roles of heterogeneities in order to endow materials with new quantum functions for energy and quantum information science applications. In this article, several representative examples are reviewed on the recent progress in connecting the heterogeneities to the quantum behaviors of real materials. Specifically, three intertwined topic areas are assessed: i) Reveal the structural, electronic, magnetic, vibrational, and optical degrees of freedom of heterogeneities. ii) Understand the effect of heterogeneities on the behaviors of quantum states in host material systems. iii) Control heterogeneities for new quantum functions. This progress is achieved by establishing the atomistic-level structure-property relationships associated with heterogeneities in quantum materials. The understanding of the interactions between electronic, magnetic, photonic, and vibrational states of heterogeneities enables the design of new quantum materials, including topological matter and quantum light emitters based on heterogenous 2D materials.