arXiv subjects

N. M. Anoop Krishnan

Publications and source records attributed to N. M. Anoop Krishnan.

51 records · Page 3Linked to original sources

arXiv2022

Learning the Dynamics of Particle-based Systems with Lagrangian Graph Neural Networks

Physical systems are commonly represented as a combination of particles, the individual dynamics of which govern the system dynamics. However, traditional approaches require the knowledge of several abstract quantities such as the energy or force to infer the dynamics of these particles. Here, we present a framework, namely, Lagrangian graph neural network (LGnn), that provides a strong inductive bias to learn the Lagrangian of a particle-based system directly from the trajectory. We test our approach on challenging systems with constraints and drag -- LGnn outperforms baselines such as feed-forward Lagrangian neural network (Lnn) with improved performance. We also show the zero-shot generalizability of the system by simulating systems two orders of magnitude larger than the trained one and also hybrid systems that are unseen by the model, a unique feature. The graph architecture of LGnn significantly simplifies the learning in comparison to Lnn with ~25 times better performance on ~20 times smaller amounts of data. Finally, we show the interpretability of LGnn, which directly provides physical insights on drag and constraint forces learned by the model. LGnn can thus provide a fillip toward understanding the dynamics of physical systems purely from observable quantities.

cs.LG↗

arXiv2022

Learning the Stress-Strain Fields in Digital Composites using Fourier Neural Operator

Increased demands for high-performance materials have led to advanced composite materials with complex hierarchical designs. However, designing a tailored material microstructure with targeted properties and performance is extremely challenging due to the innumerable design combinations and prohibitive computational costs for physics-based solvers. In this study, we employ a neural operator-based framework, namely Fourier neural operator (FNO) to learn the mechanical response of 2D composites. We show that the FNO exhibits high-fidelity predictions of the complete stress and strain tensor fields for geometrically complex composite microstructures with very few training data and purely based on the microstructure. The model also exhibits zero-shot generalization on unseen arbitrary geometries with high accuracy. Furthermore, the model exhibits zero-shot super-resolution capabilities by predicting high-resolution stress and strain fields directly from low-resolution input configurations. Finally, the model also provides high-accuracy predictions of equivalent measures for stress-strain fields allowing realistic upscaling of the results.

cond-mat.mtrl-sci↗

arXiv2022

Unsupervised Graph Neural Network Reveals the Structure--Dynamics Correlation in Disordered Systems

Learning the structure--dynamics correlation in disordered systems is a long-standing problem. Here, we use unsupervised machine learning employing graph neural networks (GNN) to investigate the local structures in disordered systems. We test our approach on 2D binary A65B35 LJ glasses and extract structures corresponding to liquid, supercooled and glassy states at different cooling rates. The neighborhood representation of atoms learned by a GNN in an unsupervised fashion, when clustered, reveal local structures with varying potential energies. These clusters exhibit dynamical heterogeneity in the structure in congruence with their local energy landscape. Altogether, the present study shows that unsupervised graph embedding can reveal the structure--dynamics correlation in disordered structures.

cond-mat.dis-nn↗

arXiv2021

Lagrangian Neural Network with Differentiable Symmetries and Relational Inductive Bias

Realistic models of physical world rely on differentiable symmetries that, in turn, correspond to conservation laws. Recent works on Lagrangian and Hamiltonian neural networks show that the underlying symmetries of a system can be easily learned by a neural network when provided with an appropriate inductive bias. However, these models still suffer from issues such as inability to generalize to arbitrary system sizes, poor interpretability, and most importantly, inability to learn translational and rotational symmetries, which lead to the conservation laws of linear and angular momentum, respectively. Here, we present a momentum conserving Lagrangian neural network (MCLNN) that learns the Lagrangian of a system, while also preserving the translational and rotational symmetries. We test our approach on linear and non-linear spring systems, and a gravitational system, demonstrating the energy and momentum conservation. We also show that the model developed can generalize to systems of any arbitrary size. Finally, we discuss the interpretability of the MCLNN, which directly provides physical insights into the interactions of multi-particle systems.

cs.LG↗

arXiv2021

MatSciBERT: A Materials Domain Language Model for Text Mining and Information Extraction

An overwhelmingly large amount of knowledge in the materials domain is generated and stored as text published in peer-reviewed scientific literature. Recent developments in natural language processing, such as bidirectional encoder representations from transformers (BERT) models, provide promising tools to extract information from these texts. However, direct application of these models in the materials domain may yield suboptimal results as the models themselves may not be trained on notations and jargon that are specific to the domain. Here, we present a materials-aware language model, namely, MatSciBERT, which is trained on a large corpus of scientific literature published in the materials domain. We further evaluate the performance of MatSciBERT on three downstream tasks, namely, abstract classification, named entity recognition, and relation extraction, on different materials datasets. We show that MatSciBERT outperforms SciBERT, a language model trained on science corpus, on all the tasks. Further, we discuss some of the applications of MatSciBERT in the materials domain for extracting information, which can, in turn, contribute to materials discovery or optimization. Finally, to make the work accessible to the larger materials community, we make the pretrained and finetuned weights and the models of MatSciBERT freely accessible.

cs.CL↗

arXiv2021

Revealing the Compositional Control of Electrical, Mechanical, Optical, and Physical Properties of Inorganic Glasses

Inorganic glasses, produced by the melt-quenching of a concoction of minerals, compounds, and elements, can possess unique optical and elastic properties along with excellent chemical, and thermal durability. Despite the ubiquitous use of glasses for critical applications such as touchscreen panels, windshields, bioactive implants, optical fibers and sensors, kitchen and laboratory glassware, thermal insulators, nuclear waste immobilization, optical lenses, and solid electrolytes, their composition-structure-property relationships remain poorly understood. Here, exploiting largescale experimental data on inorganic glasses and explainable machine learning algorithms, we develop composition-property models for twenty-five properties, which are in agreement with experimental observations. These models are further interpreted using a game-theoretic concept namely, Shapley additive explanations, to understand the role of glass components in controlling the final property. The analysis reveals that the components present in the glass, such as network formers, modifiers, and the intermediates, play distinct roles in governing each of the optical, physical, electrical, and mechanical properties of glasses. Additionally, these components exhibit interdependence, the magnitude of which is different for different properties. While the physical origins of some of these interdependencies could be attributed to known phenomena such as "boron anomaly", "mixed modifier effect", and the "Loewenstein rule", the majority of the remaining ones requires further experimental and computational analysis of the glass structure. Thus, our work paves the way for decoding the "glass genome", which can provide the recipe for discovery of novel glasses, while also shedding light into the fundamental factors governing the composition-structure-property relationships.

cond-mat.mtrl-sci↗

arXiv2021

Unveiling the Glass Veil: Elucidating the Optical Properties in Glasses with Interpretable Machine Learning

Due to their excellent optical properties, glasses are used for various applications ranging from smartphone screens to telescopes. Developing compositions with tailored Abbe number (Vd) and refractive index (nd), two crucial optical properties, is a major challenge. To this extent, machine learning (ML) approaches have been successfully used to develop composition-property models. However, these models are essentially black-box in nature and suffer from the lack of interpretability. In this paper, we demonstrate the use of ML models to predict the composition-dependent variations of Vd and n at 587.6 nm (nd). Further, using Shapely Additive exPlanations (SHAP), we interpret the ML models to identify the contribution of each of the input components toward a target prediction. We observe that the glass formers such as SiO2, B2O3, and P2O5, and intermediates like TiO2, PbO, and Bi2O3 play a significant role in controlling the optical properties. Interestingly, components that contribute toward increasing the nd are found to decrease the Vd and vice-versa. Finally, we develop the Abbe diagram, also known as the "glass veil", using the ML models, allowing accelerated discovery of new glasses for optical properties beyond the experimental pareto front. Overall, employing explainable ML, we discover the hidden compositional control on the optical properties of oxide glasses.

physics.optics↗

arXiv2021

Looking Through Glass: Knowledge Discovery from Materials Science Literature using Natural Language Processing

Most of the knowledge in materials science literature is in the form of unstructured data such as text and images. Here, we present a framework employing natural language processing, which automates text and image comprehension and precision knowledge extraction from inorganic glasses' literature. The abstracts are automatically categorized using latent Dirichlet allocation (LDA), providing a way to classify and search semantically linked publications. Similarly, a comprehensive summary of images and plots are presented using the 'Caption Cluster Plot' (CCP), which provides direct access to the images buried in the papers. Finally, we combine the LDA and CCP with the chemical elements occurring in the manuscript to present an 'Elemental map', a topical and image-wise distribution of chemical elements in the literature. Overall, the framework presented here can be a generic and powerful tool to extract and disseminate material-specific information on composition-structure-processing-property dataspaces, allowing insights into fundamental problems relevant to the materials science community and accelerated materials discovery.

cs.DL↗

arXiv2020

Scalable Gaussian Processes for Predicting the Properties of Inorganic Glasses with Large Datasets

Gaussian process regression (GPR) is a useful technique to predict composition--property relationships in glasses as the method inherently provides the standard deviation of the predictions. However, the technique remains restricted to small datasets due to the substantial computational cost associated with it. Here, using a scalable GPR algorithm, namely, kernel interpolation for scalable structured Gaussian processes (KISS-GP) along with massively scalable GP (MSGP), we develop composition--property models for inorganic glasses based on a large dataset with more than 100,000 glass compositions, 37 components, and nine important properties, namely, density, Young's, shear, and bulk moduli, thermal expansion coefficient, Vickers' hardness, refractive index, glass transition temperature, and liquidus temperature. Finally, to accelerate glass design, the models developed here are shared publicly as part of a package, namely, Python for Glass Genomics (PyGGi).

physics.comp-ph↗

arXiv2020

On the Allowable or Forbidden Nature of Vapor-Deposited Glasses

Vapor deposition can yield glasses that are more stable than those obtained by the traditional melt-quenching route. However, it remains unclear whether vapor-deposited glasses are "allowable" or "forbidden," that is, if they are equivalent to glasses formed by cooling extremely slowly a liquid or if they differ in nature from melt-quenched glasses. Here, based on reactive molecular dynamics simulation (MD) of silica glasses, we demonstrate that the allowable or forbidden nature of vapor-deposited glasses depends on the temperature of the substrate and, in turn, is found to be encoded in their medium-range order structure.

cond-mat.mtrl-sci↗

arXiv2019

Deep Learning Aided Rational Design of Oxide Glasses

Despite the extensive usage of oxide glasses for a few millennia, the composition-property relationships in these materials still remain poorly understood. While empirical and physics-based models have been used to predict properties, these remain limited to a few select compositions or a series of glasses. Designing new glasses requires a priori knowledge of how the composition of a glass dictates its properties such as stiffness, density, or processability. Thus, accelerated design of glasses for targeted applications remain impeded due to the lack of universal composition-property models. Herein, using deep learning, we present a methodology for the rational design of oxide glasses. Exploiting a large dataset of glasses comprising of up to 37 oxide components and more than 100,000 glass compositions, we develop high-fidelity deep neural networks for the prediction of eight properties that enable the design of glasses, namely, density, Young's modulus, shear modulus, hardness, glass transition temperature, thermal expansion coefficient, liquidus temperature, and refractive index. These models are by far the most extensive models developed as they cover the entire range of human-made glass compositions. We demonstrate that the models developed here exhibit excellent predictability, ensuring close agreement with experimental observations. Using these models, we develop a series of new design charts, termed as glass selection charts. These charts enable the rational design of functional glasses for targeted applications by identifying unique compositions that satisfy two or more constraints, on both compositions and properties, simultaneously. The generic design approach presented herein could catalyze machine-learning assisted materials design and discovery for a large class of materials including metals, ceramics, and proteins.

cond-mat.mtrl-sci↗

arXiv2019

Cooling Rate Effects on the Structure of 45S5 Bioglass: Computational and Experimental Evidence of Si--P Avoidance

Due to its ability to bond with living tissues upon dissolution, 45S5 bioglass and related compositions are promising materials for the replacement, regeneration, and repair of hard tissues in the human body. However, the details of their atomic structure remain unclear. This is partially due to the non-equilibrium nature of glasses, as their non-crystalline structure is highly dependent on their thermal history, namely, the cooling rate used during quenching. Herein, using molecular dynamics (MD) simulations and magic angle spinning nuclear magnetic resonance (MAS-NMR) spectroscopy experiments, we investigate the structure of the nominal 45S5 bioglass composition prepared using cooling rates ranging over several orders of magnitude. We show that the simulations results are in very good agreement with experimental data, provided that they are extrapolated toward lower cooling rates achieved in experiments. These results highlight that previously reported inconsistencies between simulations and experiments stem from the difference in cooling rate, thereby addressing one of the longstanding questions on the structure of bioglass. Based on these results, we demonstrate the existence of a Si--P avoidance behavior, which may be key in controlling the bioactivity of 45S5 bioglass.

cond-mat.mtrl-sci↗

arXiv2019

Predicting Young's Modulus of Glasses with Sparse Datasets using Machine Learning

Machine learning (ML) methods are becoming popular tools for the prediction and design of novel materials. In particular, neural network (NN) is a promising ML method, which can be used to identify hidden trends in the data. However, these methods rely on large datasets and often exhibit overfitting when used with sparse dataset. Further, assessing the uncertainty in predictions for a new dataset or an extrapolation of the present dataset is challenging. Herein, using Gaussian process regression (GPR), we predict Young's modulus for silicate glasses having sparse dataset. We show that GPR significantly outperforms NN for sparse dataset, while ensuring no overfitting. Further, thanks to the nonparametric nature, GPR provides quantitative bounds for the reliability of predictions while extrapolating. Overall, GPR presents an advanced ML methodology for accelerating the development of novel functional materials such as glasses.

cond-mat.mtrl-sci↗

arXiv2018

A New Transferable Interatomic Potential for Molecular Dynamics Simulations of Borosilicate Glasses

Borosilicate glasses are traditionally challenging to model using atomic scale simulations due to the composition and thermal history dependence of the coordination state of B atoms. Here, we report a new empirical interatomic potential that shows a good transferability over a wide range of borosilicate glasses--ranging from pure silicate to pure borate end members--while relying on a simple formulation and a constant set of energy parameters. In particular, we show that our new potential accurately predicts the compositional dependence of the average coordination number of boron atoms, glass density, overall short-range and medium-range order structure, and shear viscosity values for several borosilicate glasses and liquids. This suggests that our new potential could be used to gain new insights into the structure of a variety of advanced borosilicate glasses to help elucidate composition-structure-property relationships--including in complex nuclear waste immobilization glasses.

cond-mat.dis-nn↗

arXiv2017

Predicting the dissolution kinetics of silicate glasses using machine learning

Predicting the dissolution rates of silicate glasses in aqueous conditions is a complex task as the underlying mechanism(s) remain poorly understood and the dissolution kinetics can depend on a large number of intrinsic and extrinsic factors. Here, we assess the potential of data-driven models based on machine learning to predict the dissolution rates of various aluminosilicate glasses exposed to a wide range of solution pH values, from acidic to caustic conditions. Four classes of machine learning methods are investigated, namely, linear regression, support vector machine regression, random forest, and artificial neural network. We observe that, although linear methods all fail to describe the dissolution kinetics, the artificial neural network approach offers excellent predictions, thanks to its inherent ability to handle non-linear data. Overall, we suggest that a more extensive use of machine learning approaches could significantly accelerate the design of novel glasses with tailored properties.

cond-mat.dis-nn↗