Search

In recent years, machine learning (ML), especially graph neural network (GNN) models, has been successfully used for fast and accurate prediction of material properties. However, most ML models rely on relaxed crystal structures to develop descriptors for accurate predictions. Generating these relaxed crystal structures can be expensive and time-consuming, thus requiring an additional processing step for models that rely on them. To address this challenge, structure-agnostic methods have been developed, which use fixed-length descriptors engineered based on human knowledge about the material. However, the fixed-length descriptors are often hand-engineered and require extensive domain knowledge and generally are not used in the context of learnable models which are known to have a superior performance. Recent advancements have proposed learnable frameworks that can construct representations based on stoichiometry alone, allowing the flexibility of using deep learning frameworks as well as leveraging structure-agnostic learning. In this work, we propose three different pretraining strategies that can be used to pretrain these structure-agnostic, learnable frameworks to further improve the downstream material property prediction performance. We incorporate strategies such as self-supervised learning (SSL), fingerprint learning (FL), and multimodal learning (ML) and demonstrate their efficacy on downstream tasks for the Roost architecture, a popular structure-agnostic framework. Our results show significant improvement in small data sets and data efficiency in the larger data sets, underscoring the potential of our pretrain strategies that effectively leverage unlabeled data for accurate material property prediction.

The fast and untraceable virus mutations take lives of thousands of people before the immune system can produce the inhibitory antibody. The recent outbreak of COVID-19 infected and killed thousands of people in the world. Rapid methods in finding peptides or antibody sequences that can inhibit the viral epitopes of SARS-CoV-2 will save the life of thousands. To predict neutralizing antibodies for SARS-CoV-2 in a high-throughput manner, in this paper, we use different machine learning (ML) model to predict the possible inhibitory synthetic antibodies for SARS-CoV-2. We collected 1933 virus-antibody sequences and their clinical patient neutralization response and trained an ML model to predict the antibody response. Using graph featurization with variety of ML methods, like XGBoost, Random Forest, Multilayered Perceptron, Support Vector Machine and Logistic Regression, we screened thousands of hypothetical antibody sequences and found nine stable antibodies that potentially inhibit SARS-CoV-2. We combined bioinformatics, structural biology, and Molecular Dynamics (MD) simulations to verify the stability of the candidate antibodies that can inhibit SARS-CoV-2.

Machine learning encompasses tools and algorithms that are now becoming popular in almost all scientific and technological fields. This is true for molecular dynamics as well, where machine learning offers promises of extracting valuable information from the enormous amounts of data generated by simulation of complex systems. We provide here a review of our current understanding of goals, benefits, and limitations of machine learning techniques for computational studies on atomistic systems, focusing on the construction of empirical force fields from ab initio databases and the determination of reaction coordinates for free energy computation and enhanced sampling.

Thermal properties of materials are essential to many applications of thermal electronic devices. Density functional theory (DFT) has shown capability in obtaining an accurate calculation. However, the expensive computational cost limits the application of the DFT method for high-throughput screening of materials. Recently, machine learning models, especially graph neural networks (GNNs), have demonstrated high accuracy in many material properties' prediction, such as bandgap and formation energy, but fail to accurately predict heat capacity(C V) due to the limitation in capturing crystallographic features. In our study, we have implemented the material informatics transformer (MatInFormer) framework, which has been pretrained on lattice reconstruction tasks. This approach has shown proficiency in capturing essential crystallographic features. By concatenating these features with human-designed descriptors, we achieved a mean absolute error of 4.893 and 4.505 J/(mol K) in our predictions. Our findings underscore the efficacy of the MatInFormer framework in leveraging crystallography, augmented with additional information processing capabilities. [ABSTRACT FROM AUTHOR]

Meshing is a critical, but user-intensive process necessary for stable and accurate simulations in computational fluid dynamics (CFD). Mesh generation is often a bottleneck in CFD pipelines. Adaptive meshing techniques allow the mesh to be updated automatically to produce an accurate solution for the problem at hand. Existing classical techniques for adaptive meshing require either additional functionality out of solvers, many training simulations, or both. Current machine learning techniques often require substantial computational cost for training data generation, and are restricted in scope to the training data flow regime. Mesh Deep Q Network (MeshDQN) is developed as a general purpose deep reinforcement learning framework to iteratively coarsen meshes while preserving target property calculation. A graph neural network based deep Q network is used to select mesh vertices for removal and solution interpolation is used to bypass expensive simulations at each step in the improvement process. MeshDQN requires a single simulation prior to mesh coarsening, while making no assumptions about flow regime, mesh type, or solver, only requiring the ability to modify meshes directly in a CFD pipeline. MeshDQN successfully improves meshes for two 2D airfoils. [ABSTRACT FROM AUTHOR]

Developing fast surrogates for Partial Differential Equations (PDEs) will accelerate design and optimization in almost all scientific and engineering applications. Neural networks have been receiving ever-increasing attention and demonstrated remarkable success in computational modeling of PDEs, however; their prediction accuracy is not at the level of full deployment. In this work, we utilize the transformer architecture, the backbone of numerous state-of-the-art AI models, to learn the dynamics of physical systems as the mixing of spatial patterns learned by a convolutional autoencoder. Moreover, we incorporate the idea of multi-scale hierarchical time-stepping to increase the prediction speed and decrease accumulated error over time. Our model achieves similar or better results in predicting the time-evolution of Navier–Stokes equations compared to the powerful Fourier Neural Operator (FNO) and two transformer-based neural operators OFormer and Galerkin Transformer. The code and data are available on https://github.com/BaratiLab/MST_PDE. [ABSTRACT FROM AUTHOR]

Identification of nonlinear dynamical systems using data-driven frameworks facilitates the prediction and control of systems in a range of applications. Identification of a single system from the measurements of the system's states leads to the discovery of explicit or implicit models that cannot generalize beyond the system for which the data are provided. By learning the effect of parameters in the system, we propose a generalizable model for the Identification of Parametric forms of dynamical systems using Integer Programming (IP2). We first build general libraries of basis functions that take into account both states and parameters. Subsequently, leveraging dimension analysis and the assumption of having integer coefficients in the equations, we show that our framework can identify the exact forms of parametric mechanical dynamical systems like an ideal pendulum or an inverted pendulum on a cart. Moreover, by applying object tracking techniques and taking advantage of a sequential filtering scheme, we can identify the state and energy equations of these dynamical systems from videos of the systems, i.e. pixel space noisy data, rather than state-space measurements. The results show that using integer programming makes the proposed framework significantly (more than 40 times in the case of inverted pendulum on a cart) more robust to noise compared to previous optimization models. • Integer Programming is used to identify parametric equations of dynamical systems. • Object detection and sequential filters can process data from videos of systems. • System's parameters are considered similar to states to find full parametric equations. • Dimension analysis and integer programming make IP2 robust to noise. • IP2 is used to identify equations of systems like inverse pendulum on a cart. [ABSTRACT FROM AUTHOR]

Recent advances in machine learning (ML) based methodologies have accelerated the prediction of the physical properties of materials. These ML models, however, rely on large amounts of simulated or experimental data to make a reliable prediction. This dependence on large amounts of data can be a roadblock to building ML models since collecting the data is prohibitively expensive and time-consuming. In this work, we propose two sampling strategies to reliably train machine learning models in the lowest amounts of data. Our algorithms alleviate the need to generate large datasets to train machine learning models. We demonstrate the effectiveness of these sampling strategies by improving the performance of Crystal Graph Convolutional Neural Network (CGCNN) on four different datasets. Using the proposed strategies, we can reach the benchmark performance of CGCNN models in fewer data samples. [Display omitted] [ABSTRACT FROM AUTHOR]

Many scientific phenomena are modeled by Partial Differential Equations (PDEs). The development of data gathering tools along with the advances in machine learning (ML) techniques have raised opportunities for data-driven identification of governing equations from experimentally observed data. We propose an ML method to discover the terms involved in the equation from two-dimensional spatiotemporal data. Robust and useful physical features are extracted from data samples to represent the specific behaviors imposed by each mathematical term in the equation. Compared to the previous models, this idea provides us with the ability to discover 2D equations with time derivatives of different orders, and also to identify new underlying physics on which the model has not been trained. Moreover, the model can work with small sets of low-resolution data while avoiding instability caused by numerical differentiations. The results indicate robustness of the features extracted based on prior knowledge in comparison to automatically detected features by a Three-dimensional Convolutional Neural Network (3D CNN) given the same amounts of data. Although particular PDEs are studied in this work, the idea of the proposed approach could be extended for reliable identification of various PDEs. [Display omitted] • Data-driven methods are used to identify Partial Differential Equations. • Physical features are extracted from spatiotemporal data for identification. • A machine learning framework exploits features to form the PDE from mathematical terms. • The framework is able to discover PDEs that are not seen in the training. • Features show more robustness than detected features using deep learning methods. [ABSTRACT FROM AUTHOR]