Exploring order parameters and dynamic processes in disordered systems via variational autoencoders
arXiv:2006.10267 · doi:10.1126/sciadv.abd5084
Abstract
We suggest and implement an approach for the bottom-up description of systems undergoing large-scale structural changes and chemical transformations from dynamic atomically resolved imaging data, where only partial or uncertain data on atomic positions are available. This approach is predicated on the synergy of two concepts, the parsimony of physical descriptors and general rotational invariance of non-crystalline solids, and is implemented using a rotationally-invariant extension of the variational autoencoder applied to semantically segmented atom-resolved data seeking the most effective reduced representation for the system that still contains the maximum amount of original information. This approach allowed us to explore the dynamic evolution of electron beam-induced processes in a silicon-doped graphene system, but it can be also applied for a much broader range of atomic-scale and mesoscopic phenomena to introduce the bottom-up order parameters and explore their dynamics with time and in response to external stimuli.
arXiv commentThe revised version
References in corpus (24)
- Auto-Encoding Variational Bayes
- An Introduction to Variational Autoencoders
- Deep Learning of Atomically Resolved Scanning Transmission Electron Microscopy Images: Chemical Identification and Tracking Local Transformations
- Suppression of Octahedral Tilts and Associated Changes of Electronic Properties at Epitaxial Oxide Heterostructure Interfaces
- Spectroscopic imaging of single atoms within a bulk solid
- Stone-Wales--type transformations in carbon nanostructures driven by electron irradiation
- Manipulating low-dimensional materials down to the level of single atoms with electron irradiation
- Deep Learning Analysis of Defect and Phase Evolution During Electron Beam Induced Transformations in WS2
- Silicon-carbon bond inversions driven by 60 keV electrons in graphene
- Single atom manipulation and control in a scanning transmission electron microscope
- Imaging Atomic-Level Random Walk of a Point Defect in Graphene
- Electron-Beam Manipulation of Silicon Dopants in Graphene
- Assembling Di- and Multiatomic Si Clusters in Graphene via Electron Beam Manipulation
- Patterned Probes for High Precision 4D-STEM Bragg Measurements
- Single-atom spectroscopy of phosphorus dopants implanted into graphene
- Towards atomically precise manipulation of 2D nanostructures in the electron microscope
- Manifold Learning of Four-dimensional Scanning Transmission Electron Microscopy
- Direct atomic fabrication and dopant positioning in Si using electron beams with active real time image-based feedback
- Electron-beam Introduction of Heteroatomic Pt-Si Structures in Graphene
- Mitigating E-beam-induced Hydrocarbon Deposition on Graphene for Atomic-Scale Scanning Transmission Electron Microscopy Studies
- Explicitly disentangling image content from translation and rotation with spatial-VAE
- Toward Decoding the Relationship between Domain Structure and Functionality in Ferroelectrics via Hidden Latent Variables
- Atomic mechanisms for the Si atom dynamics in graphene: chemical transformations at the edge and in the bulk
- E-beam manipulation of Si atoms on graphene edges with aberration-corrected STEM
Cited by in corpus (11)
- AtomAI: A Deep Learning Framework for Analysis of Image and Spectroscopy Data in (Scanning) Transmission Electron Microscopy and Beyond
- Deep Learning for Automated Experimentation in Scanning Transmission Electron Microscopy
- A Physically-informed Graph-based Order Parameter for the Universal Characterization of Atomic Structures
- Machine learning of Kondo physics using variational autoencoders and symbolic regression
- Explainable Representation Learning of Small Quantum States
- Benchmark tests of atom segmentation deep learning models with a consistent dataset
- Unraveling the Impact of Initial Choices and In-Loop Interventions on Learning Dynamics in Autonomous Scanning Probe Microscopy
- Unsupervised learning of ferroic variants from atomically resolved STEM images
- Disentangling ferroelectric wall dynamics and identification of pinning mechanisms via deep learning
- Physical discovery in representation learning via conditioning on prior knowledge: applications for ferroelectric domain dynamics
- Detecting Atomic Scale Surface Defects in STM of TMDs with Ensemble Deep Learning