arXiv subjects

Maxim A. Ziatdinov

Publications and source records attributed to Maxim A. Ziatdinov.

At least 19 recordsLinked to original sources

arXiv2025

Reward driven discovery of the optimal microstructure representations with invariant variational autoencoders

Microscopy techniques generate vast amounts of complex image data that in principle can be used to discover simpler, interpretable, and parsimonious forms to reveal the underlying physical structures, such as elementary building blocks in molecular systems or order parameters and phases in crystalline materials. Variational Autoencoders (VAEs) provide a powerful means of constructing such low-dimensional representations, but their performance heavily depends on multiple non-myopic design choices, which are often optimized through trial-and-error and empirical analysis. To enable automated and unbiased optimization of VAE workflows, we investigated reward-based strategies for evaluating latent space representations. Using Piezoresponse Force Microscopy data as a model system, we examined multiple policies and reward functions that can serve as a foundation for automated optimization. Our analysis shows that approximating the latent space with Gaussian Mixture Models (GMM) and Bayesian Gaussian Mixture Models (BGMM) provides a strong basis for constructing reward functions capable of estimating model efficiency and guiding the search for optimal parsimonious representations.

cs.LG↗

arXiv2025

Integrating Predictive and Generative Capabilities by Latent Space Design via the DKL-VAE Model

We introduce a Deep Kernel Learning Variational Autoencoder (VAE-DKL) framework that integrates the generative power of a Variational Autoencoder (VAE) with the predictive nature of Deep Kernel Learning (DKL). The VAE learns a latent representation of high-dimensional data, enabling the generation of novel structures, while DKL refines this latent space by structuring it in alignment with target properties through Gaussian Process (GP) regression. This approach preserves the generative capabilities of the VAE while enhancing its latent space for GP-based property prediction. We evaluate the framework on two datasets: a structured card dataset with predefined variational factors and the QM9 molecular dataset, where enthalpy serves as the target function for optimization. The model demonstrates high-precision property prediction and enables the generation of novel out-of-training subset structures with desired characteristics. The VAE-DKL framework offers a promising approach for high-throughput material discovery and molecular design, balancing structured latent space organization with generative flexibility.

cs.LG↗

arXiv2024

Reward driven workflows for unsupervised explainable analysis of phases and ferroic variants from atomically resolved imaging data

Rapid progress in aberration corrected electron microscopy necessitates development of robust methods for the identification of phases, ferroic variants, and other pertinent aspects of materials structure from imaging data. While unsupervised methods for clustering and classification are widely used for these tasks, their performance can be sensitive to hyperparameter selection in the analysis workflow. In this study, we explore the effects of descriptors and hyperparameters on the capability of unsupervised ML methods to distill local structural information, exemplified by discovery of polarization and lattice distortion in Sm doped BiFeO3 (BFO) thin films. We demonstrate that a reward-driven approach can be used to optimize these key hyperparameters across the full workflow, where rewards were designed to reflect domain wall continuity and straightness, ensuring that the analysis aligns with the material's physical behavior. This approach allows us to discover local descriptors that are best aligned with the specific physical behavior, providing insight into the fundamental physics of materials. We further extend the reward driven workflows to disentangle structural factors of variation via optimized variational autoencoder (VAE). Finally, the importance of well-defined rewards was explored as a quantifiable measure of success of the workflow.

cond-mat.mtrl-sci↗

arXiv2024

Bayesian Co-navigation: Dynamic Designing of the Materials Digital Twins via Active Learning

Scientific advancement is universally based on the dynamic interplay between theoretical insights, modelling, and experimental discoveries. However, this feedback loop is often slow, including delayed community interactions and the gradual integration of experimental data into theoretical frameworks. This challenge is particularly exacerbated in domains dealing with high-dimensional object spaces, such as molecules and complex microstructures. Hence, the integration of theory within automated and autonomous experimental setups, or theory in the loop automated experiment, is emerging as a crucial objective for accelerating scientific research. The critical aspect is not only to use theory but also on-the-fly theory updates during the experiment. Here, we introduce a method for integrating theory into the loop through Bayesian co-navigation of theoretical model space and experimentation. Our approach leverages the concurrent development of surrogate models for both simulation and experimental domains at the rates determined by latencies and costs of experiments and computation, alongside the adjustment of control parameters within theoretical models to minimize epistemic uncertainty over the experimental object spaces. This methodology facilitates the creation of digital twins of material structures, encompassing both the surrogate model of behavior that includes the correlative part and the theoretical model itself. While demonstrated here within the context of functional responses in ferroelectric materials, our approach holds promise for broader applications, the exploration of optical properties in nanoclusters, microstructure-dependent properties in complex materials, and properties of molecular systems. The analysis code that supports the funding is publicly available at https://github.com/Slautin/2024_Co-navigation/tree/main

cond-mat.mtrl-sci↗

arXiv2024

Co-orchestration of Multiple Instruments to Uncover Structure-Property Relationships in Combinatorial Libraries

The rapid growth of automated and autonomous instrumentations brings forth an opportunity for the co-orchestration of multimodal tools, equipped with multiple sequential detection methods, or several characterization tools to explore identical samples. This can be exemplified by the combinatorial libraries that can be explored in multiple locations by multiple tools simultaneously, or downstream characterization in automated synthesis systems. In the co-orchestration approaches, information gained in one modality should accelerate the discovery of other modalities. Correspondingly, the orchestrating agent should select the measurement modality based on the anticipated knowledge gain and measurement cost. Here, we propose and implement a co-orchestration approach for conducting measurements with complex observables such as spectra or images. The method relies on combining dimensionality reduction by variational autoencoders with representation learning for control over the latent space structure, and integrated into iterative workflow via multi-task Gaussian Processes (GP). This approach further allows for the native incorporation of the system's physics via a probabilistic model as a mean function of the GP. We illustrated this method for different modalities of piezoresponse force microscopy and micro-Raman on combinatorial $Sm-BiFeO_3$ library. However, the proposed framework is general and can be extended to multiple measurement modalities and arbitrary dimensionality of measured signals. The analysis code that supports the funding is publicly available at https://github.com/Slautin/2024_Co-orchestration.

cond-mat.mtrl-sci↗

arXiv2023

Deep Kernel Methods Learn Better: From Cards to Process Optimization

The ability of deep learning methods to perform classification and regression tasks relies heavily on their capacity to uncover manifolds in high-dimensional data spaces and project them into low-dimensional representation spaces. In this study, we investigate the structure and character of the manifolds generated by classical variational autoencoder (VAE) approaches and deep kernel learning (DKL). In the former case, the structure of the latent space is determined by the properties of the input data alone, while in the latter, the latent manifold forms as a result of an active learning process that balances the data distribution and target functionalities. We show that DKL with active learning can produce a more compact and smooth latent space which is more conducive to optimization compared to previously reported methods, such as the VAE. We demonstrate this behavior using a simple cards data set and extend it to the optimization of domain-generated trajectories in physical systems. Our findings suggest that latent manifolds constructed through active learning have a more beneficial structure for optimization problems, especially in feature-rich target-poor scenarios that are common in domain sciences, such as materials synthesis, energy storage, and molecular discovery. The jupyter notebooks that encapsulate the complete analysis accompany the article.

cs.LG↗

arXiv2023

Discovery of structure-property relations for molecules via hypothesis-driven active learning over the chemical space

Discovery of the molecular candidates for applications in drug targets, biomolecular systems, catalysts, photovoltaics, organic electronics, and batteries, necessitates development of machine learning algorithms capable of rapid exploration of the chemical spaces targeting the desired functionalities. Here we introduce a novel approach for the active learning over the chemical spaces based on hypothesis learning. We construct the hypotheses on the possible relationships between structures and functionalities of interest based on a small subset of data and introduce them as (probabilistic) mean functions for the Gaussian process. This approach combines the elements from the symbolic regression methods such as SISSO and active learning into a single framework. The primary focus of constructing this framework is to approximate physical laws in an active learning regime toward a more robust predictive performance, as traditional evaluation on hold-out sets in machine learning doesn't account for out-of-distribution effects and may lead to a complete failure on unseen chemical space. Here, we demonstrate it for the QM9 dataset, but it can be applied more broadly to datasets from both domains of molecular and solid-state materials sciences.

cs.LG↗

arXiv2023

A dynamic Bayesian optimized active recommender system for curiosity-driven Human-in-the-loop automated experiments

Optimization of experimental materials synthesis and characterization through active learning methods has been growing over the last decade, with examples ranging from measurements of diffraction on combinatorial alloys at synchrotrons, to searches through chemical space with automated synthesis robots for perovskites. In virtually all cases, the target property of interest for optimization is defined apriori with limited human feedback during operation. In contrast, here we present the development of a new type of human in the loop experimental workflow, via a Bayesian optimized active recommender system (BOARS), to shape targets on the fly, employing human feedback. We showcase examples of this framework applied to pre-acquired piezoresponse force spectroscopy of a ferroelectric thin film, and then implement this in real time on an atomic force microscope, where the optimization proceeds to find symmetric piezoresponse amplitude hysteresis loops. It is found that such features appear more affected by subsurface defects than the local domain structure. This work shows the utility of human-augmented machine learning approaches for curiosity-driven exploration of systems across experimental domains. The analysis reported here is summarized in Colab Notebook for the purpose of tutorial and application to other data: https://github.com/arpanbiswas52/varTBO

cs.LG↗

arXiv2023

Deep Learning for Automated Experimentation in Scanning Transmission Electron Microscopy

Machine learning (ML) has become critical for post-acquisition data analysis in (scanning) transmission electron microscopy, (S)TEM, imaging and spectroscopy. An emerging trend is the transition to real-time analysis and closed-loop microscope operation. The effective use of ML in electron microscopy now requires the development of strategies for microscopy-centered experiment workflow design and optimization. Here, we discuss the associated challenges with the transition to active ML, including sequential data analysis and out-of-distribution drift effects, the requirements for the edge operation, local and cloud data storage, and theory in the loop operations. Specifically, we discuss the relative contributions of human scientists and ML agents in the ideation, orchestration, and execution of experimental workflows and the need to develop universal hyper languages that can apply across multiple platforms. These considerations will collectively inform the operationalization of ML in next-generation experimentation.

cond-mat.mtrl-sci↗

arXiv2023

Roadmap on Deep Learning for Microscopy

Through digital imaging, microscopy has evolved from primarily being a means for visual observation of life at the micro- and nano-scale, to a quantitative tool with ever-increasing resolution and throughput. Artificial intelligence, deep neural networks, and machine learning are all niche terms describing computational methods that have gained a pivotal role in microscopy-based research over the past decade. This Roadmap is written collectively by prominent researchers and encompasses selected aspects of how machine learning is applied to microscopy image data, with the aim of gaining scientific knowledge by improved image quality, automated detection, segmentation, classification and tracking of objects, and efficient merging of information from multiple imaging modalities. We aim to give the reader an overview of the key developments and an understanding of possibilities and limitations of machine learning for microscopy. It will be of interest to a wide cross-disciplinary audience in the physical sciences and life sciences.

physics.optics↗

arXiv2022

A roadmap for edge computing enabled automated multidimensional transmission electron microscopy

The advent of modern, high-speed electron detectors has made the collection of multidimensional hyperspectral transmission electron microscopy datasets, such as 4D-STEM, a routine. However, many microscopists find such experiments daunting since such datasets' analysis, collection, long-term storage, and networking remain challenging. Some common issues are the large and unwieldy size of the said datasets, often running into several gigabytes, non-standardized data analysis routines, and a lack of clarity about the computing and network resources needed to utilize the electron microscope fully. However, the existing computing and networking bottlenecks introduce significant penalties in each step of these experiments, and thus, real-time analysis-driven automated experimentation for multidimensional TEM is exceptionally challenging. One solution is integrating microscopy with edge computing, where moderately powerful computational hardware performs the preliminary analysis before handing off the heavier computation to HPC systems. In this perspective, we trace the roots of computation in modern electron microscopy, demonstrate deep learning experiments running on an edge system, and discuss the networking requirements for tying together microscopes, edge computers, and HPC systems.

cond-mat.mtrl-sci↗

arXiv2022

Automated Experiments of Local Non-linear Behavior in Ferroelectric Materials

We develop and implement an automated experiment in multimodal imaging to probe structural, chemical, and functional behaviors in complex materials and elucidate the dominant physical mechanisms that control device function. Here the emergence of non-linear electromechanical responses in piezoresponse force microscopy (PFM) is explored. Non-linear responses in PFM can originate from multiple mechanisms, including intrinsic material responses often controlled by domain structure, surface topography that affects the mechanical phenomena at the tip-surface junction, and, potentially, the presence of surface contaminants. Using an automated experiment to probe the origins of non-linear behavior in model ferroelectric lead titanate (PTO) and ferroelectric Al0.93B0.07N films, it was found that PTO showed asymmetric nonlinear behavior across a/c domain walls and a broadened high nonlinear response region around c/c domain walls. In contrast, for Al0.93B0.07N, well-poled regions showed high linear piezoelectric responses paired with low non-linear responses and regions that were multidomain indicated low linear responses and high nonlinear responses. We show that formulating dissimilar exploration strategies in deep kernel learning as alternative hypotheses allows for establishing the preponderant physical mechanisms behind the non-linear behaviors, suggesting that this approach automated experiments can potentially discern between competing physical mechanisms. This technique can also be extended to electron, probe, and chemical imaging.

cond-mat.mtrl-sci↗

arXiv2022

Bayesian Optimization in Continuous Spaces via Virtual Process Embeddings

Automated chemical synthesis, materials fabrication, and spectroscopic physical measurements often bring forth the challenge of process trajectory optimization, i.e., discovering the time dependence of temperature, electric field, or pressure that gives rise to optimal properties. Due to the high dimensionality of the corresponding vectors, these problems are not directly amenable to Bayesian Optimization (BO). Here we propose an approach based on the combination of the generative statistical models, specifically variational autoencoders, and Bayesian optimization. Here, the set of potential trajectories is formed based on best practices in the field, domain intuition, or human expertise. The variational autoencoder is used to encode the thus generated trajectories as a latent vector, and also allows for the generation of trajectories via sampling from latent space. In this manner, Bayesian Optimization of the process is realized in the latent space of the system, reducing the problem to a low-dimensional one. Here we apply this approach to a ferroelectric lattice model and demonstrate that this approach allows discovering the field trajectories that maximize curl in the system. The analysis of the corresponding polarization and curl distributions allows the relevant physical mechanisms to be decoded.

cond-mat.dis-nn↗

arXiv2022

Physical discovery in representation learning via conditioning on prior knowledge: applications for ferroelectric domain dynamics

Recent advances in electron, scanning probe, optical, and chemical imaging and spectroscopy yield bespoke data sets containing the information of structure and functionality of complex systems. In many cases, the resulting data sets are underpinned by low-dimensional simple representations encoding the factors of variability within the data. The representation learning methods seek to discover these factors of variability, ideally further connecting them with relevant physical mechanisms. However, generally the task of identifying the latent variables corresponding to actual physical mechanisms is extremely complex. Here, we explore an approach based on conditioning the data on the known (continuous) physical parameters, and systematically compare it with the previously introduced approach based on the invariant variational autoencoders. The conditional variational autoencoders (cVAE) approach does not rely on the existence of the invariant transforms, and hence allows for much greater flexibility and applicability. Interestingly, cVAE allows for limited extrapolation outside of the original domain of the conditional variable. However, this extrapolation is limited compared to the cases when true physical mechanisms are known, and the physical factor of variability can be disentangled in full. We further show that introducing the known conditioning results in the simplification of the latent distribution if the conditioning vector is correlated with the factor of variability in the data, thus allowing to separate relevant physical factors. We initially demonstrate this approach using 1D and 2D examples on a synthetic dataset and then extend it to the analysis of experimental data on ferroelectric domain dynamics visualized via Piezoresponse Force Microscopy.

cond-mat.mtrl-sci↗

arXiv2021

Experimental discovery of structure-property relationships in ferroelectric materials via active learning

Emergent functionalities of structural and topological defects in ferroelectric materials underpin an extremely broad spectrum of applications ranging from domain wall electronics to high dielectric and electromechanical responses. Many of these have been discovered and quantified via local scanning probe microscopy methods. However, the search for these functionalities has until now been based by either trial and error or using auxiliary information such as topography or domain wall structure to identify potential objects of interest based on the intuition of operator or preexisting hypotheses, with subsequent manual exploration. Here, we report the development and implementation of a machine learning framework that actively discovers relationships between local domain structure and polarization switching characteristics in ferroelectric materials encoded in the hysteresis loop. The latter and descriptors such as nucleation bias, coercive bias, hysteresis loop area, or more complex functionals of hysteresis loop shape and corresponding uncertainties are used to guide the discovery via automated piezoresponse force microscopy (PFM) and spectroscopy experiments. As such, this approach combines the power of machine learning methods to learn the correlative relationships between high dimensional data, and human-based physics insights encoded in the acquisition function. For ferroelectric, this automated workflow demonstrates that the discovery path and sampling points of on-field and off-field hysteresis loops are largely different, indicating the on-field and off-field hysteresis loops are dominated by different mechanisms. The proposed approach is universal and can be applied to a broad range of modern imaging and spectroscopy methods ranging from other scanning probe microscopy modalities to electron microscopy and chemical imaging.

cond-mat.mtrl-sci↗

arXiv2021

Automated and Autonomous Experiment in Electron and Scanning Probe Microscopy

Machine learning and artificial intelligence (ML/AI) are rapidly becoming an indispensable part of physics research, with domain applications ranging from theory and materials prediction to high-throughput data analysis. In parallel, the recent successes in applying ML/AI methods for autonomous systems from robotics through self-driving cars to organic and inorganic synthesis are generating enthusiasm for the potential of these techniques to enable automated and autonomous experiment (AE) in imaging. Here, we aim to analyze the major pathways towards AE in imaging methods with sequential image formation mechanisms, focusing on scanning probe microscopy (SPM) and (scanning) transmission electron microscopy ((S)TEM). We argue that automated experiments should necessarily be discussed in a broader context of the general domain knowledge that both informs the experiment and is increased as the result of the experiment. As such, this analysis should explore the human and ML/AI roles prior to and during the experiment, and consider the latencies, biases, and knowledge priors of the decision-making process. Similarly, such discussion should include the limitations of the existing imaging systems, including intrinsic latencies, non-idealities and drifts comprising both correctable and stochastic components. We further pose that the role of the AE in microscopy is not the exclusion of human operators (as is the case for autonomous driving), but rather automation of routine operations such as microscope tuning, etc., prior to the experiment, and conversion of low latency decision making processes on the time scale spanning from image acquisition to human-level high-order experiment planning.

cs.LG↗

arXiv2021

Mapping causal patterns in crystalline solids

The evolution of the atomic structures of the combinatorial library of Sm-substituted thin film BiFeO3 along the phase transition boundary from the ferroelectric rhombohedral phase to the non-ferroelectric orthorhombic phase is explored using scanning transmission electron microscopy (STEM). Localized properties including polarization, lattice parameter, and chemical composition are parameterized from atomic-scale imaging and their causal relationships are reconstructed using a linear non-Gaussian acyclic model (LiNGAM). This approach is further extended toward exploring the spatial variability of the causal coupling using the sliding window transform method, which revealed that new causal relationships emerged both at the expected locations, such as domain walls and interfaces, but also at additional regions forming clusters in the vicinity of the walls or spatially distributed features. While the exact physical origins of these relationships are unclear, they likely represent nanophase separated regions in the morphotropic phase boundaries. Overall, we pose that an in-depth understanding of complex disordered materials away from thermodynamic equilibrium necessitates understanding not only of the generative processes that can lead to observed microscopic states, but also the causal links between multiple interacting subsystems.

physics.data-an↗

arXiv2018

Learning from imperfections: constructing phase diagrams from atomic imaging of fluctuations

Materials characterization and property measurements are a cornerstone of material science, providing feedback from synthesis to applications. Traditionally, a single sample is used to derive information on a single point in composition space, and imperfections, impurities and stochastic details of material structure are deemed irrelevant or complicating factors in analysis. Here we demonstrate that atomic-scale studies of a single nominal composition can provide information on a finite area of chemical space. This information can be used to reconstruct the material properties in a finite composition and temperature range. We develop a statistical physics-based framework that incorporates chemical and structural data to infer effective atomic interactions driving segregation in a La5/8Ca3/8MnO3 thin-film. A variational autoencoder is used to determine anomalous behaviors in the composition phase diagram. This study provides a framework for creating generative models from diverse data and provides direct insight into the driving forces for cation segregation in manganites.

cond-mat.mtrl-sci↗