APL Machine Learning最新文献_第4页

Leveraging graph neural networks and neural operator techniques for high-fidelity mesh-based physics simulations 利用图神经网络和神经算子技术进行高保真网格物理模拟

APL Machine Learning Pub Date : 2023-11-22 DOI: 10.1063/5.0167014

Zeqing Jin, Bowen Zheng, Changgon Kim, Grace X. Gu

{"title":"Leveraging graph neural networks and neural operator techniques for high-fidelity mesh-based physics simulations","authors":"Zeqing Jin, Bowen Zheng, Changgon Kim, Grace X. Gu","doi":"10.1063/5.0167014","DOIUrl":"https://doi.org/10.1063/5.0167014","url":null,"abstract":"Developing fast and accurate computational models to simulate intricate physical phenomena has been a persistent research challenge. Recent studies have demonstrated remarkable capabilities in predicting various physical outcomes through machine learning-assisted approaches. However, it remains challenging to generalize current methods, usually crafted for a specific problem, to other more complex or broader scenarios. To address this challenge, we developed graph neural network (GNN) models with enhanced generalizability derived from the distinct GNN architecture and neural operator techniques. As a proof of concept, we employ our GNN models to predict finite element (FE) simulation results for three-dimensional solid mechanics problems with varying boundary conditions. Results show that our GNN model achieves accurate and robust performance in predicting the stress and deformation profiles of structures compared with FE simulations. Furthermore, the neural operator embedded GNN approach enables learning and predicting various solid mechanics problems in a generalizable fashion, making it a promising approach for surrogate modeling.","PeriodicalId":502250,"journal":{"name":"APL Machine Learning","volume":"20 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139246860","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A cloud platform for sharing and automated analysis of raw data from high throughput polymer MD simulations 共享和自动分析高通量聚合物 MD 模拟原始数据的云平台

APL Machine Learning Pub Date : 2023-11-17 DOI: 10.1063/5.0160937

T. Xie, Ha-Kyung Kwon, D. Schweigert, Sheng Gong, A. France-Lanord, A. Khajeh, E. Crabb, Michael Puzon, Chris Fajardo, Will Powelson, Y. Shao-horn, Jeffrey C. Grossman

{"title":"A cloud platform for sharing and automated analysis of raw data from high throughput polymer MD simulations","authors":"T. Xie, Ha-Kyung Kwon, D. Schweigert, Sheng Gong, A. France-Lanord, A. Khajeh, E. Crabb, Michael Puzon, Chris Fajardo, Will Powelson, Y. Shao-horn, Jeffrey C. Grossman","doi":"10.1063/5.0160937","DOIUrl":"https://doi.org/10.1063/5.0160937","url":null,"abstract":"Open material databases storing thousands of material structures and their properties have become the cornerstone of modern computational materials science. Yet, the raw simulation outputs are generally not shared due to their huge size. In this work, we describe a cloud-based platform to enable fast post-processing of the trajectories and to facilitate sharing of the raw data. As an initial demonstration, our database includes 6286 molecular dynamics trajectories for amorphous polymer electrolytes (5.7 terabytes of data). We create a public analysis library at https://github.com/TRI-AMDD/htp_md to extract ion transport properties from the raw data using expert-designed functions and machine learning models. The analysis is run automatically on the cloud, and the results are uploaded onto an open database. Our platform encourages users to contribute both new trajectory data and analysis functions via public interfaces. Finally, we create a front-end user interface at https://www.htpmd.matr.io/ for browsing and visualization of our data. We envision the platform to be a new way of sharing raw data and new insights for the materials science community.","PeriodicalId":502250,"journal":{"name":"APL Machine Learning","volume":"7 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139265884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Calibration in machine learning uncertainty quantification: Beyond consistency to target adaptivity 机器学习不确定性量化中的校准：从一致性到目标适应性

APL Machine Learning Pub Date : 2023-09-12 DOI: 10.1063/5.0174943

Pascal Pernot

引用次数: 0

Experiment-based deep learning approach for power allocation with a programmable metasurface 基于实验的深度学习方法，利用可编程元面进行功率分配

APL Machine Learning Pub Date : 2023-07-26 DOI: 10.1063/5.0184328

Jingxin Zhang, J. Xi, Peixing Li, Ray C. C. Cheung, A. Wong, Jensen Li

{"title":"Experiment-based deep learning approach for power allocation with a programmable metasurface","authors":"Jingxin Zhang, J. Xi, Peixing Li, Ray C. C. Cheung, A. Wong, Jensen Li","doi":"10.1063/5.0184328","DOIUrl":"https://doi.org/10.1063/5.0184328","url":null,"abstract":"Metasurfaces designed with deep learning approaches have emerged as efficient tools for manipulating electromagnetic waves to achieve beam steering and power allocation objectives. However, the effects of complex environmental factors like obstacle blocking and other unavoidable scattering need to be sufficiently considered for practical applications. In this work, we employ an experiment-based deep learning approach for programmable metasurface design to control powers delivered to specific locations generally with obstacle blocking. Without prior physical knowledge of the complex system, large sets of experimental data can be efficiently collected with a programmable metasurface to train a deep neural network (DNN). The experimental data can inherently incorporate complex factors that are difficult to include if only simulation data are used for training. Moreover, the DNN can be updated by collecting new experimental data on-site to adapt to changes in the environment. Our proposed experiment-based DNN demonstrates significant potential for intelligent wireless communication, imaging, sensing, and quiet-zone control for practical applications.","PeriodicalId":502250,"journal":{"name":"APL Machine Learning","volume":"62 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139354347","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Hyena neural operator for partial differential equations 偏微分方程的鬣狗神经算子

APL Machine Learning Pub Date : 2023-06-28 DOI: 10.1063/5.0177276

Saurabh Patil, Zijie Li, Amir Barati Farimani

{"title":"Hyena neural operator for partial differential equations","authors":"Saurabh Patil, Zijie Li, Amir Barati Farimani","doi":"10.1063/5.0177276","DOIUrl":"https://doi.org/10.1063/5.0177276","url":null,"abstract":"Numerically solving partial differential equations typically requires fine discretization to resolve necessary spatiotemporal scales, which can be computationally expensive. Recent advances in deep learning have provided a new approach to solving partial differential equations that involves the use of neural operators. Neural operators are neural network architectures that learn mappings between function spaces and have the capability to solve partial differential equations based on data. This study utilizes a novel neural operator called Hyena, which employs a long convolutional filter that is parameterized by a multilayer perceptron. The Hyena operator is an operation that enjoys sub-quadratic complexity and enjoys a global receptive field at the meantime. This mechanism enhances the model’s comprehension of the input’s context and enables data-dependent weight for different partial differential equation instances. To measure how effective the layers are in solving partial differential equations, we conduct experiments on the diffusion–reaction equation and Navier–Stokes equation and compare it with the Fourier neural operator. Our findings indicate that the Hyena neural operator can serve as an efficient and accurate model for learning the partial differential equation solution operator. The data and code used can be found at https://github.com/Saupatil07/Hyena-Neural-Operator.","PeriodicalId":502250,"journal":{"name":"APL Machine Learning","volume":"217 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139368012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0