arXiv subjects

Maxim Ziatdinov

Publications and source records attributed to Maxim Ziatdinov.

At least 55 records · Page 3Linked to original sources

arXiv2022

Optimizing Training Trajectories in Variational Autoencoders via Latent Bayesian Optimization Approach

Unsupervised and semi-supervised ML methods such as variational autoencoders (VAE) have become widely adopted across multiple areas of physics, chemistry, and materials sciences due to their capability in disentangling representations and ability to find latent manifolds for classification and regression of complex experimental data. Like other ML problems, VAEs require hyperparameter tuning, e.g., balancing the Kullback Leibler (KL) and reconstruction terms. However, the training process and resulting manifold topology and connectivity depend not only on hyperparameters, but also their evolution during training. Because of the inefficiency of exhaustive search in a high-dimensional hyperparameter space for the expensive to train models, here we explored a latent Bayesian optimization (zBO) approach for the hyperparameter trajectory optimization for the unsupervised and semi-supervised ML and demonstrate for joint-VAE with rotational invariances. We demonstrate an application of this method for finding joint discrete and continuous rotationally invariant representations for MNIST and experimental data of a plasmonic nanoparticles material system. The performance of the proposed approach has been discussed extensively, where it allows for any high dimensional hyperparameter tuning or trajectory optimization of other ML models.

cs.LG↗

arXiv2022

Exploring Physics of Ferroelectric Domain Walls in Real Time: Deep Learning Enabled Scanning Probe Microscopy

The functionality of ferroelastic domain walls in ferroelectric materials is explored in real-time via the in-situ implementation of computer vision algorithms in scanning probe microscopy (SPM) experiment. The robust deep convolutional neural network (DCNN) is implemented based on a deep residual learning framework (Res) and holistically-nested edge detection (Hed), and ensembled to minimize the out-of-distribution drift effects. The DCNN is implemented for real-time operations on SPM, converting the data stream into the semantically segmented image of domain walls and the corresponding uncertainty. We further demonstrate the pre-selected experimental workflows on thus discovered domain walls, and report alternating high- and low- polarization dynamic (out-of-plane) ferroelastic domain walls in a (PbTiO3) PTO thin film and high polarization dynamic (out-of-plane) at short ferroelastic walls (compared with long ferroelastic walls) in a lead zirconate titanate (PZT) thin film. This work establishes the framework for real-time DCNN analysis of data streams in scanning probe and other microscopies and highlights the role of out-of-distribution effects and strategies to ameliorate them in real time analytics.

cond-mat.mtrl-sci↗

arXiv2022

Unsupervised learning of ferroic variants from atomically resolved STEM images

An approach for the analysis of atomically resolved scanning transmission electron microscopy data with multiple ferroic variants in the presence of imaging non-idealities and chemical variabilities based on a rotationally invariant variational autoencoder (rVAE) is presented. We show that an optimal local descriptor for the analysis is a sub-image centered at specific atomic units, since materials and microscope distortions preclude the use of an ideal lattice as a reference point. The applicability of unsupervised clustering and dimensionality reduction methods is explored and are shown to produce clusters dominated by chemical and microscope effects, with a large number of classes required to establish the presence of rotational variants. Comparatively, the rVAE allows extraction of the angle corresponding to the orientation of ferroic variants explicitly, enabling straightforward identification of the ferroic variants as regions with constant or smoothly changing latent variables and sharp orientational changes. This approach allows further exploration of the chemical variability by separating the rotational degrees of freedom via rVAE and searching for remaining variability in the system. The code used in the manuscript is available at https://github.com/saimani5/ferroelectric_domains_rVAE.

cond-mat.mtrl-sci↗

arXiv2022

Hypothesis Learning in Automated Experiment: Application to Combinatorial Materials Libraries

Machine learning is rapidly becoming an integral part of experimental physical discovery via automated and high-throughput synthesis, and active experiments in scattering and electron/probe microscopy. This, in turn, necessitates the development of active learning methods capable of exploring relevant parameter spaces with the smallest number of steps. Here we introduce an active learning approach based on co-navigation of the hypothesis and experimental spaces. This is realized by combining the structured Gaussian Processes containing probabilistic models of the possible system's behaviors (hypotheses) with reinforcement learning policy refinement (discovery). This approach closely resembles classical human-driven physical discovery, when several alternative hypotheses realized via models with adjustable parameters are tested during an experiment. We demonstrate this approach for exploring concentration-induced phase transitions in combinatorial libraries of Sm-doped BiFeO3 using Piezoresponse Force Microscopy, but it is straightforward to extend it to higher-dimensional parameter spaces and more complex physical problems once the experimental workflow and hypothesis-generation are available.

cond-mat.mtrl-sci↗

arXiv2022

Automated experiment in 4D-STEM: exploring emergent physics and structural behaviors

Automated experiments in 4D Scanning Transmission Electron Microscopy are implemented for rapid discovery of local structures, symmetry-breaking distortions, and internal electric and magnetic fields in complex materials. Deep kernel learning enables active learning of the relationship between local structure and a 4D-STEM based descriptors. With this, efficient and "intelligent" probing of dissimilar structural elements to discover desired physical functionality is made possible. This approach allows effective navigation of the sample in an automated fashion guided by either a pre-determined physical phenomenon, such as strongest electric field magnitude, or in an exploratory fashion. We verify the approach first on pre-acquired 4D-STEM data, and further implement it experimentally on an operational STEM. The experimental discovery workflow is demonstrated using graphene, and subsequently extended towards a lesser-known layered 2D van der Waal material, MnPS3. This approach establishes a paradigm for physics-driven automated 4D-STEM experiments that enable probing the physics of strongly correlated systems and quantum materials and devices, as well as exploration of beam sensitive materials.

cond-mat.mtrl-sci↗

arXiv2022

Physics is the New Data

The rapid development of machine learning (ML) methods has fundamentally affected numerous applications ranging from computer vision, biology, and medicine to accounting and text analytics. Until now, it was the availability of large and often labeled data sets that enabled significant breakthroughs. However, the adoption of these methods in classical physical disciplines has been relatively slow, a tendency that can be traced to the intrinsic differences between correlative approaches of purely data-based ML and the causal hypothesis-driven nature of physical sciences. Furthermore, anomalous behaviors of classical ML necessitate addressing issues such as explainability and fairness of ML. We also note the sequence in which deep learning became mainstream in different scientific disciplines - starting from medicine and biology and then towards theoretical chemistry, and only after that, physics - is rooted in the progressively more complex level of descriptors, constraints, and causal structures available for incorporation in ML architectures. Here we put forth that over the next decade, physics will become a new data, and this will continue the transition from dot-coms and scientific computing concepts of the 90ies to big data of 2000-2010 to deep learning of 2010-2020 to physics-enabled scientific ML.

physics.data-an↗

arXiv2022

Active learning in open experimental environments: selecting the right information channel(s) based on predictability in deep kernel learning

Active learning methods are rapidly becoming the integral component of automated experiment workflows in imaging, materials synthesis, and computation. The distinctive aspect of many experimental scenarios is the presence of multiple information channels, including both the intrinsic modalities of the measurement system and the exogenous environment and noise signals. One of the key tasks in experimental studies is hence establishing which of these channels is predictive of the behaviors of interest. Here we explore the problem of discovery of the optimal predictive channel for structure-property relationships (in microscopy) using deep kernel learning for modality selection in an active experiment setting. We further pose that this approach can be directly applicable to similar active learning tasks in automated synthesis and the discovery of quantitative structure-activity relations in molecular systems.

cs.LG↗

arXiv2022

Discovering Invariant Spatial Features in Electron Energy Loss Spectroscopy Images on the Mesoscopic and Atomic Levels

Over the last two decades, Electron Energy Loss Spectroscopy (EELS) imaging with a scanning transmission electron microscope (STEM) has emerged as a technique of choice for visualizing complex chemical, electronic, plasmonic, and phononic phenomena in complex materials and structures. The availability of the EELS data necessitates the development of methods to analyze multidimensional datasets with complex spatial and energy structures. Traditionally, the analysis of these data sets has been based on analysis of individual spectra, one at a time, whereas the spatial structure and correlations between individual spatial pixels containing the relevant information of the physics of underpinning processes have generally been ignored and analyzed only via the visualization as 2D maps. Here we develop a machine learning-based approach and workflows for the analysis of spatial structures in 3D EELS data sets using a combination of dimensionality reduction and multichannel rotationally-invariant variational autoencoders. This approach is illustrated for the analysis of both the plasmonic phenomena in a system of nanowires and in the core excitations in functional oxides using low loss and core loss EELS, respectively. The code developed in this manuscript is open sourced and freely available and provided as a Jupyter notebook for the interested reader here.

cond-mat.mtrl-sci↗

arXiv2021

Bridging microscopy with molecular dynamics and quantum simulations: An AtomAI based pipeline

Recent advances in (scanning) transmission electron microscopy have enabled routine generation of large volumes of high-veracity structural data on 2D and 3D materials, naturally offering the challenge of using these as starting inputs for atomistic simulations. In this fashion, theory will address experimentally emerging structures, as opposed to the full range of theoretically possible atomic configurations. However, this challenge is highly non-trivial due to the extreme disparity between intrinsic time scales accessible to modern simulations and microscopy, as well as latencies of microscopy and simulations per se. Addressing this issue requires as a first step bridging the instrumental data flow and physics-based simulation environment, to enable the selection of regions of interest and exploring them using physical simulations. Here we report the development of the machine learning workflow that directly bridges the instrument data stream into Python-based molecular dynamics and density functional theory environments using pre-trained neural networks to convert imaging data to physical descriptors. The pathways to ensure the structural stability and compensate for the observational biases universally present in the data are identified in the workflow. This approach is used for a graphene system to reconstruct optimized geometry and simulate temperature-dependent dynamics including adsorption of Cr as an ad-atom and graphene healing effects. However, it is universal and can be used for other material systems.

physics.data-an↗

arXiv2021

Exploring causal physical mechanisms via non-gaussian linear models and deep kernel learning: applications for ferroelectric domain structures

Rapid emergence of the multimodal imaging in scanning probe, electron, and optical microscopies have brought forth the challenge of understanding the information contained in these complex data sets, targeting both the intrinsic correlations between different channels and further exploring the underpinning causal physical mechanisms. Here, we develop such analysis framework for the Piezoresponse Force Microscopy. We argue that under certain conditions, we can bootstrap experimental observations with the prior knowledge of materials structure to get information on certain non-observed properties, and demonstrate linear causal analysis for PFM observables. We further demonstrate that this approach can be extended to complex descriptors using the deep kernel learning (DKL) model. In this DKL analysis, we use the prior information on domain structure within the image to predict the physical properties. This analysis demonstrates the correlative relationships between morphology, piezoresponse, elastic property, etc. at nanoscale. The prediction of morphology and other physical parameters illustrates a mutual interaction between surface condition and physical properties in ferroelectric materials. This analysis is universal and can be extended to explore the correlative relationships of other multi-channel datasets.

cond-mat.mtrl-sci↗

arXiv2021

Identification and Correction of Temporal and Spatial Distortions in Scanning Transmission Electron Microscopy

Scanning transmission electron microscopy (STEM) has become the technique of choice for quantitative characterization of atomic structure of materials, where the minute displacements of atomic columns from high-symmetry positions can be used to map strain, polarization, octahedra tilts, and other physical and chemical order parameter fields. The latter can be used as inputs into mesoscopic and atomistic models, providing insight into the correlative relationships and generative physics of materials on the atomic level. However, these quantitative applications of STEM necessitate understanding the microscope induced image distortions and developing the pathways to compensate them both as part of a rapid calibration procedure for in situ imaging, and the post-experimental data analysis stage. Here, we explore the spatiotemporal structure of the microscopic distortions in STEM using multivariate analysis of the atomic trajectories in the image stacks. Based on the behavior of principal component analysis (PCA), we develop the Gaussian process (GP)-based regression method for quantification of the distortion function. The limitations of such an approach and possible strategies for implementation as a part of in-line data acquisition in STEM are discussed. The analysis workflow is summarized in a Jupyter notebook that can be used to retrace the analysis and analyze the reader's data.

cond-mat.mtrl-sci↗

arXiv2021

Describing condensed matter from atomically resolved imaging data: from structure to generative and causal models

The development of high-resolution imaging methods such as electron and scanning probe microscopy and atomic probe tomography have provided a wealth of information on structure and functionalities of solids. The availability of this data in turn necessitates development of approaches to derive quantitative physical information, much like the development of scattering methods in the early XX century which have given one of the most powerful tools in condensed matter physics arsenal. Here, we argue that this transition requires adapting classical macroscopic definitions, that can in turn enable fundamentally new opportunities in understanding physics and chemistry. For example, many macroscopic definitions such as symmetry can be introduced locally only in a Bayesian sense, balancing the prior knowledge of materials' physics and experimental data to yield posterior probability distributions. At the same time, a wealth of local data allows fundamentally new approaches for the description of solids based on construction of statistical and physical generative models, akin to Ginzburg-Landau thermodynamic models. Finally, we note that availability of observational data opens pathways towards exploring causal mechanisms underpinning solid structure and functionality.

cond-mat.mtrl-sci↗

arXiv2021

Physics makes the difference: Bayesian optimization and active learning via augmented Gaussian process

Both experimental and computational methods for the exploration of structure, functionality, and properties of materials often necessitate the search across broad parameter spaces to discover optimal experimental conditions and regions of interest in the image space or parameter space of computational models. The direct grid search of the parameter space tends to be extremely time-consuming, leading to the development of strategies balancing exploration of unknown parameter spaces and exploitation towards required performance metrics. However, classical Bayesian optimization strategies based on the Gaussian process (GP) do not readily allow for the incorporation of the known physical behaviors or past knowledge. Here we explore a hybrid optimization/exploration algorithm created by augmenting the standard GP with a structured probabilistic model of the expected system's behavior. This approach balances the flexibility of the non-parametric GP approach with a rigid structure of physical knowledge encoded into the parametric model. The fully Bayesian treatment of the latter allows additional control over the optimization via the selection of priors for the model parameters. The method is demonstrated for a noisy version of the classical objective function used to evaluate optimization algorithms and further extended to physical lattice models. This methodology is expected to be universally suitable for injecting prior knowledge in the form of physical models and past data in the Bayesian optimization framework

physics.comp-ph↗

arXiv2021

Multi-objective Bayesian optimization of ferroelectric materials with interfacial control for memory and energy storage applications

Optimization of materials performance for specific applications often requires balancing multiple aspects of materials functionality. Even for the cases where generative physical model of material behavior is known and reliable, this often requires search over multidimensional parameter space to identify low-dimensional manifold corresponding to required Pareto front. Here we introduce the multi-objective Bayesian Optimization (MOBO) workflow for the ferroelectric/anti-ferroelectric performance optimization for memory and energy storage applications based on the numerical solution of the Ginzburg-Landau equation with electrochemical or semiconducting boundary conditions. MOBO is a low computational cost optimization tool for expensive multi-objective functions, where we update posterior surrogate Gaussian process models from prior evaluations, and then select future evaluations from maximizing an acquisition function. Using the parameters for a prototype bulk antiferroelectric (PbZrO3), we first develop a physics-driven decision tree of target functions from the loop structures. We further develop a physics-driven MOBO architecture to explore multidimensional parameter space and build Pareto-frontiers by maximizing two target functions jointly: energy storage and loss. This approach allows for rapid initial materials and device parameter selection for a given application and can be further expanded towards the active experiment setting. The associated notebooks provide both the tutorial on MOBO and allow to reproduce the reported analyses and apply them to other systems (https://github.com/arpanbiswas52/MOBO_AFI_Supplements).

cond-mat.mtrl-sci↗

arXiv2021

Towards Automating Structural Discovery in Scanning Transmission Electron Microscopy

Scanning transmission electron microscopy (STEM) is now the primary tool for exploring functional materials on the atomic level. Often, features of interest are highly localized in specific regions in the material, such as ferroelectric domain walls, extended defects, or second phase inclusions. Selecting regions to image for structural and chemical discovery via atomically resolved imaging has traditionally proceeded via human operators making semi-informed judgements on sampling locations and parameters. Recent efforts at automation for structural and physical discovery have pointed towards the use of "active learning" methods that utilize Bayesian optimization with surrogate models to quickly find relevant regions of interest. Yet despite the potential importance of this direction, there is a general lack of certainty in selecting relevant control algorithms and how to balance a priori knowledge of the material system with knowledge derived during experimentation. Here we address this gap by developing the automated experiment workflows with several combinations to both illustrate the effects of these choices and demonstrate the tradeoffs associated with each in terms of accuracy, robustness, and susceptibility to hyperparameters for structural discovery. We discuss possible methods to build descriptors using the raw image data and deep learning based semantic segmentation, as well as the implementation of variational autoencoder based representation. Furthermore, each workflow is applied to a range of feature sizes including NiO pillars within a La:SrMnO$_3$ matrix, ferroelectric domains in BiFeO$_3$, and topological defects in graphene. The code developed in this manuscript are open sourced and will be released at github.com/creangnc/AE_Workflows.

cond-mat.mtrl-sci↗

arXiv2021

Finding simplicity: unsupervised discovery of features, patterns, and order parameters via shift-invariant variational autoencoders

Recent advances in scanning tunneling and transmission electron microscopies (STM and STEM) have allowed routine generation of large volumes of imaging data containing information on the structure and functionality of materials. The experimental data sets contain signatures of long-range phenomena such as physical order parameter fields, polarization and strain gradients in STEM, or standing electronic waves and carrier-mediated exchange interactions in STM, all superimposed onto scanning system distortions and gradual changes of contrast due to drift and/or mis-tilt effects. Correspondingly, while the human eye can readily identify certain patterns in the images such as lattice periodicities, repeating structural elements, or microstructures, their automatic extraction and classification are highly non-trivial and universal pathways to accomplish such analyses are absent. We pose that the most distinctive elements of the patterns observed in STM and (S)TEM images are similarity and (almost-) periodicity, behaviors stemming directly from the parsimony of elementary atomic structures, superimposed on the gradual changes reflective of order parameter distributions. However, the discovery of these elements via global Fourier methods is non-trivial due to variability and lack of ideal discrete translation symmetry. To address this problem, we develop shift-invariant variational autoencoders (shift-VAE) that allow disentangling characteristic repeating features in the images, their variations, and shifts inevitable for random sampling of image space. Shift-VAEs balance the uncertainty in the position of the object of interest with the uncertainty in shape reconstruction. This approach is illustrated for model 1D data, and further extended to synthetic and experimental STM and STEM 2D data.

cond-mat.dis-nn↗

arXiv2021

Autonomous Experiments in Scanning Probe Microscopy and Spectroscopy: Choosing Where to Explore Polarization Dynamics in Ferroelectrics

Polarization dynamics in ferroelectric materials are explored via the automated experiment in Piezoresponse Force Spectroscopy. A Bayesian Optimization framework for imaging is developed and its performance for a variety of acquisition and pathfinding functions is explored using previously acquired data. The optimized algorithm is then deployed on an operational scanning probe microscope (SPM) for finding areas of large electromechanical response in a thin film of PbTiO3, with metrics showing gains of ~3x in the sampling efficiency. This approach opens the pathway to perform more complex spectroscopies in SPM that were previously not possible due to time constraints and sample stability, tip wear, and/or stochastic sample damage that occurs at specific, a priori unknown spatial positions. Potential improvements to the framework to enable the incorporation of more prior information and improve efficiency further are discussed.

cond-mat.dis-nn↗

arXiv2021

High-Throughput Study of Antisolvents on the Stability of Multicomponent Metal Halide Perovskites through Robotics-Based Synthesis and Machine Learning Approaches

Antisolvent crystallization methods are frequently used to fabricate high-quality perovskite thin films, to produce sizable single crystals, and to synthesize nanoparticles at room temperature. However, a systematic exploration of the effect of specific antisolvents on the intrinsic stability of multicomponent metal halide perovskites has yet to be demonstrated. Here, we develop a high-throughput experimental workflow that incorporates chemical robotic synthesis, automated characterization, and machine learning techniques to explore how the choice of antisolvent affects the intrinsic stability of binary perovskite systems in ambient conditions over time. Different combinations of the endmembers, MAPbI3, MAPbBr3, FAPbI3, FAPbBr3, CsPbI3, and CsPbBr3, are used to synthesize 15 combinatorial libraries, each with 96 unique combinations. In total, roughly 1100 different compositions are synthesized. Each library is fabricated twice using two different antisolvents: toluene and chloroform. Once synthesized, photoluminescence spectroscopy is automatically performed every 5 minutes for approximately 6 hours. Non-negative Matrix Factorization (NMF) is then utilized to map the time- and compositional-dependent optoelectronic properties. Through the utilization of this workflow for each library, we demonstrate that the selection of antisolvent is critical to the stability of metal halide perovskites in ambient conditions. We explore possible dynamical processes, such as halide segregation, responsible for either the stability or eventual degradation as caused by the choice of antisolvent. Overall, this high-throughput study demonstrates the vital role that antisolvents play in the synthesis of high-quality multicomponent metal halide perovskite systems.

physics.app-ph↗