Browse
Submit Data
Databases
API
Help

Dataset Information

0 Views

0 Connections

0 Citations

0 Reanalyses

0 Downloads

Omics score: 0

Deep-learning electronic-structure calculation of magnetic superstructures.

ABSTRACT: Ab initio studies of magnetic superstructures are indispensable to research on emergent quantum materials, but are currently bottlenecked by the formidable computational cost. Here, to break this bottleneck, we have developed a deep equivariant neural network framework to represent the density functional theory Hamiltonian of magnetic materials for efficient electronic-structure calculation. A neural network architecture incorporating a priori knowledge of fundamental physical principles, especially the nearsightedness principle and the equivariance requirements of Euclidean and time-reversal symmetries ([Formula: see text]), is designed, which is critical to capture the subtle magnetic effects. Systematic experiments on spin-spiral, nanotube and moiré magnets were performed, making the challenging study of magnetic skyrmions feasible.

SUBMITTER: Li H

PROVIDER: S-EPMC10766521 | biostudies-literature | 2023 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Json Xml

Publications

Deep-learning electronic-structure calculation of magnetic superstructures.

Li He H Tang Zechen Z Gong Xiaoxun X Zou Nianlong N Duan Wenhui W Xu Yong Y

Nature computational science 20230426 4

Ab initio studies of magnetic superstructures are indispensable to research on emergent quantum materials, but are currently bottlenecked by the formidable computational cost. Here, to break this bottleneck, we have developed a deep equivariant neural network framework to represent the density functional theory Hamiltonian of magnetic materials for efficient electronic-structure calculation. A neural network architecture incorporating a priori knowledge of fundamental physical principles, especi ...[more]

PMID: 38177932

Similar Datasets

Generalizing deep learning electronic structure calculation to the plane-wave basis.

Project description:Deep neural networks capable of representing the density functional theory (DFT) Hamiltonian as a function of material structure hold great promise for revolutionizing future electronic structure calculations. However, a notable limitation of previous neural networks is their compatibility solely with the atomic-orbital (AO) basis, excluding the widely used plane-wave (PW) basis. Here we overcome this critical limitation by proposing an accurate and efficient real-space reconstruction method for directly computing AO Hamiltonian matrices from PW DFT results. The reconstruction method is orders of magnitude faster than traditional projection-based methods to convert PW results to the AO basis, and the reconstructed Hamiltonian matrices can faithfully reproduce the PW electronic structure, thus bridging the longstanding gap between the AO basis deep learning electronic structure approach and PW DFT. Advantages of the PW methods, such as high accuracy, high flexibility and wide applicability, thus can be all integrated into deep learning electronic structure methods without sacrificing these methods' inherent benefits. This allows for the construction of large-scale and high-fidelity training datasets with the help of PW DFT results towards the development of precise and broadly applicable deep learning electronic structure models.

| S-EPMC11499277 | biostudies-literature

Deep-learning density functional theory Hamiltonian for efficient ab initio electronic-structure calculation.

Project description:The marriage of density functional theory (DFT) and deep-learning methods has the potential to revolutionize modern computational materials science. Here we develop a deep neural network approach to represent the DFT Hamiltonian (DeepH) of crystalline materials, aiming to bypass the computationally demanding self-consistent field iterations of DFT and substantially improve the efficiency of ab initio electronic-structure calculations. A general framework is proposed to deal with the large dimensionality and gauge (or rotation) covariance of the DFT Hamiltonian matrix by virtue of locality, and this is realized by a message-passing neural network for deep learning. High accuracy, high efficiency and good transferability of the DeepH method are generally demonstrated for various kinds of material system and physical property. The method provides a solution to the accuracy-efficiency dilemma of DFT and opens opportunities to explore large-scale material systems, as evidenced by a promising application in the study of twisted van der Waals materials.

| S-EPMC11499279 | biostudies-literature

Learning Scalable Deep Kernels with Recurrent Structure.

Project description:Many applications in speech, robotics, finance, and biology deal with sequential data, where ordering matters and recurrent structures are common. However, this structure cannot be easily captured by standard kernel functions. To model such structure, we propose expressive closed-form kernel functions for Gaussian processes. The resulting model, GP-LSTM, fully encapsulates the inductive biases of long short-term memory (LSTM) recurrent networks, while retaining the non-parametric probabilistic advantages of Gaussian processes. We learn the properties of the proposed kernels by optimizing the Gaussian process marginal likelihood using a new provably convergent semi-stochastic gradient procedure, and exploit the structure of these kernels for scalable training and prediction. This approach provides a practical representation for Bayesian LSTMs. We demonstrate state-of-the-art performance on several benchmarks, and thoroughly investigate a consequential autonomous driving application, where the predictive uncertainties provided by GP-LSTM are uniquely valuable.

| S-EPMC6334642 | biostudies-other

Deep learning-based structure segmentation and intramuscular fat annotation on lumbar magnetic resonance imaging.

Project description:BackgroundLumbar disc herniation (LDH) is a prevalent cause of low back pain. LDH patients commonly experience paraspinal muscle atrophy and fatty infiltration (FI), which further exacerbates the symptoms of low back pain. Magnetic resonance imaging (MRI) is crucial for assessing paraspinal muscle condition. Our study aims to develop a dual-model for automated muscle segmentation and FI annotation on MRI, assisting clinicians evaluate LDH conditions comprehensively.MethodsThe study retrospectively collected data diagnosed with LDH from December 2020 to May 2022. The dataset was split into a 7:3 ratio for training and testing, with an external test set prepared to validate model generalizability. The model's performance was evaluated using average precision (AP), recall and F1 score. The consistency was assessed using the Dice similarity coefficient (DSC) and Cohen's Kappa. The mean absolute percentage error (MAPE) was calculated to assess the error of the model measurements of relative cross-sectional area (rCSA) and FI. Calculate the MAPE of FI measured by threshold algorithms to compare with the model.ResultsA total of 417 patients being evaluated, comprising 216 males and 201 females, with a mean age of 49 ± 15 years. In the internal test set, the muscle segmentation model achieved an overall DSC of 0.92 ± 0.10, recall of 92.60%, and AP of 0.98. The fat annotation model attained a recall of 91.30%, F1 Score of 0.82, and Cohen's Kappa of 0.76. However, there was a decrease on the external test set. For rCSA measurements, except for longissimus (10.89%), the MAPE of other muscles was less than 10%. When comparing the errors of FI for each paraspinal muscle, the MAPE of the model was lower than that of the threshold algorithm.ConclusionThe models demonstrate outstanding performance, with lower error in FI measurement compared to thresholding algorithms.

| S-EPMC11406510 | biostudies-literature

High Level Electronic Structure Calculation of Molecular Solid-State NMR Shielding Constants.

Project description:In this work, we present a quantum mechanics/molecular mechanics (QM/MM) approach for the computation of solid-state nuclear magnetic resonance (SS-NMR) shielding constants (SCs) for molecular crystals. Besides applying standard-DFT functionals like GGAs (PBE), meta-GGAs (TPSS), and hybrids (B3LYP), we apply a double-hybrid (DSD-PBEP86) functional as well as MP2, using the domain-based local pair natural orbital (DLPNO) formalism, to calculate the NMR SCs of six amino acid crystals. All the electronic structure methods used exhibit good correlation of the NMR shieldings with respect to experimental chemical shifts for both 1H and 13C. We also find that local electronic structure is much more important than the long-range electrostatic effects for these systems, implying that cluster approaches using all-electron/Gaussian basis set methods might offer great potential for predictive computations of solid-state NMR parameters for organic solids.

| S-EPMC9009078 | biostudies-literature

Automatic calculation of myocardial perfusion reserve using deep learning with uncertainty quantification

Project description: Not available

| S-EPMC10722070 | biostudies-literature

Reconfigurable Mechanical Anisotropy in Self-Assembled Magnetic Superstructures.

Project description:Enhancement of mechanical properties in self-assembled superstructures of magnetic nanoparticles is a new emerging aspect of their remarkable collective behavior. However, how magnetic interactions modulate mechanical properties is, to date, not fully understood. Through a comprehensive Monte Carlo investigation, this study demonstrates how the mechanical properties of self-assembled magnetic nanocubes can be controlled intrinsically by the nanoparticle magnetocrystalline anisotropy (MA), as well as by the superstructure shape anisotropy, without any need for changes in structural design (i.e., nanoparticle size, shape, and packing arrangement). A low MA-to-dipolar energy ratio, as found in iron oxide and permalloy systems, favors isotropic mechanical superstructure stabilization, whereas a high ratio yields magnetically blocked nanoparticle macrospins which can give rise to metastable superferromagnetism, as expected in cobalt ferrite simple cubic supercrystals. Such full parallel alignment of the particle moments is shown to induce mechanical anisotropy, where the superior high-strength axis can be remotely reconfigured by means of an applied magnetic field. The new concepts developed here pave the way for the experimental realization of smart magneto-micromechanical systems (based, e.g., on the permanent super-magnetostriction effect illustrated here) and inspire new design rules for applied functional materials.

| S-EPMC8061348 | biostudies-literature

Electronic structure, elasticity, Debye temperature and anisotropy of cubic WO3 from first-principles calculation.

Project description:The electron structure, elastic constant, Debye temperature and anisotropy of elastic wave velocity for cubic WO3 are studied using CASTEP based on density functional theory. The optimized structure is consistent with previous work and the band gap is obtained by computing the electronic structure; the top of the valence band is not at the same point as the bottom of the conduction band, which is an indirect band-gap oxide. Electronic properties are studied from the calculation of band structure, densities of states and charge densities. The bulk and shear moduli, Young's modulus, hardness and Poisson's ratio for WO3 are studied by the elastic constants. We calculated acoustic wave velocities in different directions and estimated the Debye temperature from the acoustic velocity. The anisotropy of WO3 was analysed from the point of view of a pure wave and quasi wave.

| S-EPMC6030277 | biostudies-literature

Bipartite electronic superstructures in the vortex core of Bi2Sr2CaCu2O8+δ.

Project description:The central issue in the physics of cuprate superconductivity is the mutual relationship among superconductivity, pseudogap and broken-spatial-symmetry states. A magnetic field B suppresses superconductivity, providing an opportunity to investigate the competition among these states. Although various B-induced electronic superstructures have been reported, their energy, spatial and momentum-space structures are unclear. Here, we show using spectroscopic-imaging scanning tunnelling microscopy on Bi2Sr2CaCu2O8+δ that there are two distinct B-induced electronic superstructures, both being localized in the vortex core but appearing at different energies. In the low-energy range where the nodal Bogoliubov quasiparticles are well-defined, we observe the so-called vortex checkerboard that we identify as the B-enhanced quasiparticle interference pattern. By contrast, in the high-energy region where the pseudogap develops, the broken-spatial-symmetry patterns that pre-exist at B=0 T is locally enhanced in the vortex core. This evidences the competition between superconductivity and the broken-spatial-symmetry state that is associated with the pseudogap.

| S-EPMC4895017 | biostudies-literature

Magnetic Hamiltonian parameter estimation using deep learning techniques.

Project description:Understanding spin textures in magnetic systems is extremely important to the spintronics and it is vital to extrapolate the magnetic Hamiltonian parameters through the experimentally determined spin. It can provide a better complementary link between theories and experimental results. We demonstrate deep learning can quantify the magnetic Hamiltonian from magnetic domain images. To train the deep neural network, we generated domain configurations with Monte Carlo method. The errors from the estimations was analyzed with statistical methods and confirmed the network was successfully trained to relate the Hamiltonian parameters with magnetic structure characteristics. The network was applied to estimate experimentally observed domain images. The results are consistent with the reported results, which verifies the effectiveness of our methods. On the basis of our study, we anticipate that the deep learning techniques make a bridge to connect the experimental and theoretical approaches not only in magnetism but also throughout any scientific research.

| S-EPMC7518863 | biostudies-literature

OmicsDI is part of the ELIXIR infrastructure

OmicsDI is an Elixir interoperability service. Learn more ›

Tweets

OmicsDI Databases

PRIDE
PeptideAtlas
MassIVE
JPOST Repository
Physiome Model Repository

EGA
EVA
ENA
LINCS
PAXDB
Cell Collective

MetaboLights
Metabolomics Workbench
MetabolomeExpress
GNPS
BioModels
FAIRDOMHub

ArrayExpress
dbGaP
ExpressionAtlas
GEO
NODE

Information

Databases
Help
API
Contact us
Code on GitHub
Terms of use
Submit Data