Machine Learning In Materials Science Research Articles

Evaluating and improving the predictive accuracy of mixing enthalpies and volumes in disordered alloys from universal pretrained machine learning potentials

The advent of machine learning in materials science opens the way for exciting and ambitious simulations of large systems and long time scales with the accuracy of calculations. Recently, several pretrained universal machine learned interatomic potentials (UPMLIPs) have been published, i.e., potentials distributed with a single set of weights trained to target systems across a very wide range of chemistries and atomic arrangements. These potentials raise the hope of reducing the computational cost and methodological complexity of performing simulations compared to models that require for-purpose training. However, the application of these models needs critical evaluation to assess their usability across material types and properties. In this work, we investigate the application of the following UPMLIPs: MACE, CHGNET, and M3GNET to the context of alloy theory. We calculate the mixing enthalpies and volumes of 21 binary alloy systems and compare the results with DFT calculations to assess the performance of these potentials over different properties and types of materials. We find that the small relative energies necessary to correctly predict mixing energies are generally not reproduced by these methods with sufficient accuracy to describe correct mixing behaviors. However, the performance can be significantly improved by supplementing the training data with relevant training data. The potentials can also be used to partially accelerate these calculations by replacing the structural relaxation step. Published by the American Physical Society 2024

A novel approach for tool-narayanaswamy-moynihan model parameter extraction using multi-scale neural model

The accurate determination of parameters in the Tool-Narayanaswamy-Moynihan (TNM) model, which describes the viscoelastic behavior of glass-forming materials, is crucial for predicting material responses through various thermal histories.Traditional methods rely heavily on curve-fitting techniques; however, these often fail due to noise in the data. Furthermore, traditional methods are computationally intensive and prone to inaccuracies, particularly when dealing with complex datasets or when the initial parameter guesses are far from optimal; also, they require a skilled personnel.In this study, we propose the application of a multi-scale convolutional neural network (MCNN) as a machine learning approach to address these challenges. The MCNN model is trained on a comprehensive simulated dataset encompassing a wide range of TNM parameters, allowing it to learn intricate patterns and dependencies within the data that are difficult to capture with conventional methods. Our results show that the MCNN significantly improves the accuracy of the parameter estimations for β and x across the entire spectrum of tested conditions, achieving performance that is not only comparable to, but often surpasses, traditional curve-fitting methods. Furthermore, the MCNN demonstrates superior robustness when initial parameter estimates are suboptimal or when the dataset exhibits significant noise. Although the prediction accuracy for the activation energy Δh∗ and the pre-exponential factor log(A) was somewhat lower, the method still provides valuable estimates that can be refined with supplementary techniques.This work highlights the potential of machine learning approaches like MCNN to revolutionize the parameter extraction process in complex physical models, reducing the reliance on manual curve-fitting and providing a more automated, scalable solution. We also analyze the primary sources of prediction errors in the MCNN outputs and offer insights into future improvements, including model architecture refinements and the integration of additional physical constraints. Our findings suggest that this approach can be extended to other domains where similar models are employed, paving the way for broader applications of machine learning in materials science.

Machine-Learning Predictions of Critical Temperatures from Chemical Compositions of Superconductors.

In the quest for advanced superconducting materials, the accurate prediction of critical temperatures (Tc) poses a formidable challenge, largely due to the complex interdependencies between superconducting properties and the chemical and structural characteristics of a given material. To address this challenges, we have developed a machine-learning framework that aims to elucidate these complicated and hitherto poorly understood structure-property and property-property relationships. This study introduces a novel machine-learning-based workflow, termed the Gradient Boosted Feature Selection (GBFS), which has been tailored to predict Tc for superconductors by employing a distributed gradient-boosting framework. This approach integrates exploratory data analyses, statistical evaluations, and multicollinearity reduction techniques to select highly relevant features from a high-dimensional feature space, derived solely from the chemical composition of materials. Our methodology was rigorously tested on a data set comprising approximately 16,400 chemical compounds with around 12,000 unique chemical compositions. The GBFS workflow enabled the development of a classification model that distinguishes compositions likely to exhibit Tc values greater than 10 K. This model achieved a weighted average F1-score of 0.912, an AUC-ROC of 0.986, and an average precision score of 0.919. Additionally, the GBFS workflow underpinned a regression model that predicted Tc values with an R2 of 0.945, an MAE of 3.54 K, and an RMSE of 6.57 K on a test set obtained via random splitting. Further exploration was conducted through out-of-sample Tc predictions, particularly those exceeding the liquid nitrogen temperature, and out-of-distribution predictions for (Ca1-xLax)FeAs2 based on varying lanthanum content. The outcome of our study underscores the significance of systematic feature analysis and selection in enhancing predictive model performance, offering various advantages over models that rely primarily on algorithmic complexity. This research not only advances the field of superconductivity but also sets a precedent for the application of machine learning in materials science.

Applications of machine learning in materials science: From a methodological point of view

Discovering new functional materials with certain properties using high-throughout methods is of vital importance for materials science. The advancement of the machine learning method has provided a new, more efficient pathway for this procedure. Traditional trial-and-error methods have been supplanted by in-silico simulations, which facilitate rapid material discovery. Machine learning (ML) has further accelerated this process by uncovering patterns and relationships within complex computational data sets, thus enhancing predictive abilities for material properties and performance. This integration of computational methods and machine learning holds great potential, promising a future where material limitations are overcome, catalyzing technological breakthroughs. Furthermore, ML also inspired advancements in fields like crystallography and metallurgy, and enhancing energy storage materials. Despite challenges in model interpretability, overfitting, and data quality, ML presents an exciting evolution toward a data-driven discipline. Ultimately, ML promises to reduce the time from materials discovery to deployment, turning material limitations into catalysts for innovation.

Machine Learning Techniques for Predicting Conductive Properties of New Materials

The study "Machine Learning Techniques for Predicting Conductive Properties of New Materials" explores the application of advanced machine learning (ML) algorithms to predict the conductive properties of novel materials, accelerating the discovery and development process in materials science. Traditional methods for assessing conductive properties are often time-consuming and expensive, necessitating a more efficient approach. This research leverages various ML techniques, including supervised learning algorithms such as support vector machines, decision trees, and neural networks, to analyze large datasets of material properties and predict conductivity with high accuracy. Feature selection and engineering processes are employed to identify the most significant attributes influencing conductivity. The study also compares the performance of different ML models, optimizing hyperparameters to enhance prediction reliability. Results demonstrate that ML models can significantly reduce the experimental burden, offering rapid and precise predictions that align closely with empirical data. The integration of ML in materials science presents a transformative approach, enabling faster identification of promising conductive materials, thereby fostering advancements in electronics, energy storage, and other technological domains. The study highlights the potential of ML to revolutionize material property prediction, paving the way for accelerated innovation and application in various industries.

Artificial Intelligence and Machine Learning in Materials Science

Abstract This article provides a brief overview of the many ways that artificial intelligence and machine learning are being used for materials and manufacturing research. Several case studies show how the discovery, development, and deployment of novel materials are being dramatically accelerated through automation and data-driven models.

Prediction and Elucidation of Physical Properties of Polycrystalline Materials Using Multichannel Machine Learning of Electron Backscattering Diffraction

AbstractThe application of machine learning in materials science has yielded several benefits, including the prediction of physical properties and the improvement of experimental efficiency. However, with complex models, such as convolutional neural networks (CNN), learning has become a black box, from which no universal physical knowledge can be obtained. In this study, a highly accurate prediction of the electrical properties of polycrystalline semiconductor thin films is achieved by learning multichannel CNN models from electron backscattering diffraction (EBSD) data, that is band contrasts, grain boundaries, and inverse pole figures. In addition, it examines how the CNN model learned the correlation between the crystallinity, grain boundaries, crystallographic orientation, and carrier mobility by polarizing certain EBSD data and checking the predicted changes in carrier mobility. Physical parameters affecting carrier mobility can be extracted, which is challenging via human image recognition. The methods proposed in this study will not only enable the prediction of electrical properties from EBSD data for all materials but also will contribute to the discovery of complex physical phenomena beyond the limits of human analysis.

Predicting sound transmission loss in particle-reinforced polymer composites using machine learning: A comparative study with experimental and theoretical results

In this study, the use of machine learning (ML) models to predict sound transmission loss (STL) in particle-reinforced polymer composites has been examined. The work included extensive literature research and data extraction for training various ML models, focusing on their effectiveness in accurately predicting STL, which is crucial for evaluating acoustic performance. The method involves processing data and applying it to different ML algorithms, with the models calibrated and tested for reliability. A key aspect is comparing these models' predictions with actual experimental results and theoretical models based on the mass law in acoustics. The findings reveal the potential and limitations of ML in materials science, showing their accuracy in predicting STL and comparing them with traditional theories. This research advances the use of data-driven methods in developing and assessing acoustic materials, significantly impacting materials science and machine learning.

Autonomous generation of single photon emitting materials.

The utilization of machine learning in Materials Science underscores the critical importance of the quality and quantity of data in training models effectively. Unlike fields such as image processing and natural language processing, there is limited availability of atomistic datasets, leading to biases in training data. Particularly in the domain of materials discovery, there exists an issue of continuity in atomistic datasets. Experimental data sourced from literature and patents is usually only available for favorable data, resulting in bias in the training dataset. This study focuses on developing a SMILES-based model for generating synthetic datasets of quantum materials using a variational autoencoder. This study centers on the generation of a synthetic dataset of quantum materials specifically for quantum sensing applications, with a focus on two-level quantum molecules that exhibit a dipole blockade. The proposed technique offers an improved sampling algorithm by incorporating newly generated data into the sampling algorithm to create a more normally distributed dataset. Through this technique, the study was able to generate over 1 000 000 candidate quantum materials from a small dataset of only 8000 materials. The generated dataset identified several iodine-containing molecules as promising single photon emitting materials for potential quantum sensing applications.

Exploring the role of machine learning in materials science and engineering

Exploring the role of machine learning in materials science and engineering In this paper, Professor Dane Morgan and Research Scientist Ryan Jacobs, from the University of Wisconsin, Madison, discuss their adventures in the field of machine learning in the areas of materials science and engineering. This paper gives a very brief and inevitably biased overview of machine learning (ML) in Materials Science and Engineering (MS&E), with examples taken from our own work with collaborators. We hope it conveys our excitement about the extraordinary potential of this new area of research. MS&E focuses on developing materials with desired properties. It has led to materials innovations that underlie much of modern society, from the transistors in computers to the batteries in cars and smartphones. In recent decades, major advances in algorithms, computing power, and data access have made ML tools extremely powerful.

Explanatory Machine Learning Accelerates the Design of Graphene-Reinforced Aluminium Matrix Composites with Superior Performance

Addressing the exceptional properties of aluminium alloy composites reinforced with graphene, this study presents an interpretable machine learning approach to aid in the rapid and efficient design of such materials. Initially, data on these composites were gathered and optimised in order to create a dataset of composition/process-property. Several machine learning algorithms were used to train various models. The SHAP method was used to interpret and select the best performing model, which happened to be the CatBoost model. The model achieved accurate predictions of hardness and tensile strength, with coefficients of determination of 0.9597 and 0.9882, respectively, and average relative errors of 6.02% and 5.01%, respectively. The results obtained from the SHAP method unveiled the correlation between the composition, process and properties of aluminium alloy composites reinforced with graphene. By comparing the predicted and experimental data in this study, all machine learning models exhibited prediction errors within 10%, confirming their ability to generalise. This study offers valuable insights and support for designing high-performance aluminium matrix composites reinforced with graphene and showcases the implementation of machine learning in materials science.

(Invited) Machine Learning for Solid-State Batteries – Progress Versus Hype

Machine learning in materials science and solid-state batteries are two topics that have captured the imagination of researchers in recent years. Therefore, it is unsurprising that many researchers have attempted to apply the advances in ML to the discovery and study of materials for solid-state batteries. In this talk, I will discuss the importance of going back to fundamentals in the application of ML to materials for solid-state batteries, in particular, the critical solid electrolyte component and its interfaces. I will highlight areas where ML has had a transformative impact on our understanding and discovery of solid electrolytes, and what are some of the remaining challenges that remain to be surmounted.

Foundry-ML: a Platform for FAIR Machine Learning in Materials Science.

Foundry-ML: a Platform for FAIR Machine Learning in Materials Science.

Data quantity governance for machine learning in materials science.

Data-driven machine learning (ML) is widely employed in the analysis of materials structure-activity relationships, performance optimization and materials design due to its superior ability to reveal latent data patterns and make accurate prediction. However, because of the laborious process of materials data acquisition, ML models encounter the issue of the mismatch between a high dimension of feature space and a small sample size (for traditional ML models) or the mismatch between model parameters and sample size (for deep-learning models), usually resulting in terrible performance. Here, we review the efforts for tackling this issue via feature reduction, sample augmentation and specific ML approaches, and show that the balance between the number of samples and features or model parameters should attract great attention during data quantity governance. Following this, we propose a synergistic data quantity governance flow with the incorporation of materials domain knowledge. After summarizing the approaches to incorporating materials domain knowledge into the process of ML, we provide examples of incorporating domain knowledge into governance schemes to demonstrate the advantages of the approach and applications. The work paves the way for obtaining the required high-quality data to accelerate materials design and discovery based on ML.

Machine learning for materials science: Barriers to broader adoption

Machine learning is on a bit of a tear right now, with advances that are infiltrating nearly every aspect of our lives. In the domain of materials science, this wave seems to be growing into a tsunami. Yet, there are still real hurdles that we face to maximize its benefit. This Matter of Opinion, crafted as a result of a workshop hosted by researchers at Sandia National Laboratories and attended by a cadre of luminaries, briefly summarizes our perspective on these barriers. By recognizing these problems in a community forum, we can share the burden of their resolution together with a common purpose and coordinated effort. Machine learning is on a bit of a tear right now, with advances that are infiltrating nearly every aspect of our lives. In the domain of materials science, this wave seems to be growing into a tsunami. Yet, there are still real hurdles that we face to maximize its benefit. This Matter of Opinion, crafted as a result of a workshop hosted by researchers at Sandia National Laboratories and attended by a cadre of luminaries, briefly summarizes our perspective on these barriers. By recognizing these problems in a community forum, we can share the burden of their resolution together with a common purpose and coordinated effort.

News Article

Possible impacts of Large Language Models on X-ray spectroscopy (March 17, 2023). Recent announcement of releasing OpenAI's Generative Pretrained Transformer 4 (GPT-4) could mark the beginning of a new era of the practical application of artificial intelligence. Compared with the previous versions of OpenAI's Large Language Models (LLM), its function in handling the images even in the text-based chat system is surprising. In the presentation, which has been archived as a Youtube video (https://www.youtube.com/live/outcGtbnMuQ?feature=share), the chat system can handle a screenshot image of a handwriting note with some sketch, and can instantly transform it into an html code. Then one can create a new Web page. In addition, the reasoning ability has been upgraded considerably from the previous version, which had not been very good; previous ChatGPT seems to give the correct answer only to 26% of the junior high school level mathematics problems (see, for example, Simon Frieder et al., “Mathematical Capabilities of ChatGPT”, https://arxiv.org/abs/2301.13867). The new version seems to be much better and is now quite good at the tax calculation and the bar exam problem. Generally, the ability of text-based job has become further better than ever. The number of characters that can be handled has increased, and overall processing has become higher quality. One of the recent interesting discussions on the application of artificial intelligence to spectroscopy has been described in the article, by Jerome Workman, Jr., and Howard Mark, “Artificial Intelligence in Analytical Spectroscopy, Part I: Basic Concepts and Discussion”, Spectroscopy, 38, (2), 13–22 (2023). https://doi.org/10.56530/spectroscopy.og4284z8. Some of the related contents have been also discussed in the podcast, Analytically Speaking Podcast Episode 9, “Automating Advanced Chemometric Methods for Data Processing” (https://www.spectroscopyonline.com/view/ep-9-automating-advanced-chemometric-methods-for-data-processing). As the present artificial intelligence is more powerful in handling billions of images, for a while, image-based scientific applications could be preceding (see, for example, Lei Zhang and Shaofeng Shao, “Image-based machine learning for materials science”, Journal of Applied Physics 132, 100701 (2022). https://doi.org/10.1063/5.0087381; Samantha Phan and Christine K. Luscombe, “Recent trends in marine microplastic modeling and machine learning tools: Potential for long-term microplastic monitoring”, Journal of Applied Physics 133, 020701 (2023). https://doi.org/10.1063/5.0126358). On the same day of the announcement of GPT-4, Anthropic Inc, which is a rival company founded by former Open AI employees, released Claude. It has been tested in private systems on systems such as Notion, Quora, and DuckDuckGo, and has been very well received. The LLM and its application such as chatbots will be used more frequently than before. On the other hand, the system is like a black box, and therefore we will need to be careful when considering some real use (see, for example, Noah Giansiracusa, “Three Easy Ways to Make AI Chatbots Safer”, Scientific American, March 17, 2023, https://www.scientificamerican.com/article/three-easy-ways-to-make-ai-chatbots-safer/).

Machine Learning Enabled Image Analysis of Time-Temperature Sensing Colloidal Arrays.

Smart, responsive materials are required in various advanced applications ranging from anti-counterfeiting to autonomous sensing. Colloidal crystals are a versatile material class for optically based sensing applications owing to their photonic stopband. A careful combination of materials synthesis and colloidal mesostructure rendered such systems helpful in responding to stimuli such as gases, humidity, or temperature. Here, an approach is demonstrated to simultaneously and independently measure the time and temperature solely based on the inherent material properties of complex colloidal crystal mixtures. An array of colloidal crystals, each featuring unique film formation kinetics, is fabricated. Combined with machine learning-enabled image analysis, the colloidal crystal arrays can autonomously record isothermal heating events - readout proceeds by acquiring photographs of the applied sensor using a standard smartphone camera. The concept shows how the progressing use of machine learning in materials science has the potential to allow non-classical forms of data acquisition and evaluation. This can provide novel insights into multiparameter systems and simplify applications of novelmaterials.

Machine learning accelerates the materials discovery

As the big data generated by the development of modern experiments and computing technology becomes more and more accessible, the material design method based on machine learning (ML) has opened a new paradigm for materials science research. With its ability to automatically solve complex tasks, machine learning is being used as a new method to help discover the relevance of materials, understand materials' properties, and accelerate the discovery of materials. This paper first introduces the general process of machine learning in materials science. Secondly, the applications of machine learning in material properties prediction, classification and identification, auxiliary micro-scale characterization, phase transformation research and phase diagram construction, process optimization, service behavior evaluation, accelerating the development of computational simulation technology, multi-objective optimization and inverse design of materials are reviewed. Finally, we discuss the main challenges and possible solutions in machine learning, and predict the potential research directions.

Functional Output Regression for Machine Learning in Materials Science.

In recent years, there has been a rapid growth in the use of machine learning in material science. Conventionally, a trained predictive model describes a scalar output variable, such as thermodynamic, electronic, or mechanical properties, as a function of input descriptors that vectorize the compositional or structural features of any given material, such as molecules, chemical compositions, or crystalline systems. In machine learning of material data, on the other hand, the output variable is often given as a function. For example, when predicting the optical absorption spectrum of a molecule, the output variable is a spectral function defined in the wavelength domain. Alternatively, in predicting the microstructure of a polymer nanocomposite, the output variable is given as an image from an electron microscope, which can be represented as a two- or three-dimensional function in the image coordinate system. In this study, we consider two unified frameworks to handle such multidimensional or functional output regressions, which are applicable to a wide range of predictive analyses in material science. The first approach employs generative adversarial networks, which are known to exhibit outstanding performance in various computer vision tasks such as image generation, style transfer, and video generation. We also present another type of statistical modeling inspired by a statistical methodology referred to as functional data analysis. This is an extension of kernel regression to deal with functional outputs, and its simple mathematical structure makes it effective in modeling even with small amounts of data. We demonstrate the proposed methods through several case studies in materials science.

Image-based machine learning for materials science

Materials research studies are dealing with a large number of images, which can now be facilitated via image-based machine learning techniques. In this article, we review recent progress of machine learning-driven image recognition and analysis for the materials and chemical domains. First, the image-based machine learning that facilitates the property prediction of chemicals or materials is discussed. Second, the analysis of nanoscale images including those from a scanning electron microscope and a transmission electron microscope is discussed, which is followed by the discussion about the identification of molecular structures via image recognition. Subsequently, the image-based machine learning works to identify and classify various practical materials such as metal, ceramics, and polymers are provided, and the image recognition for a range of real-scenario device applications such as solar cells is provided in detail. Finally, suggestions and future outlook for image-based machine learning for classification and prediction tasks in the materials and chemical science are presented. This article highlights the importance of the integration of the image-based machine learning method into materials and chemical science and calls for a large-scale deployment of image-based machine learning methods for prediction and classification of images in materials and chemical science.