Pub Date : 2024-07-26DOI: 10.1088/2632-2153/ad6832
Barak Hadad, Omry Oren, A. Bahabad
Attention layers are a crucial component in many modern deep learning models, particularly those used in natural language processing and computer vision. Attention layers have been shown to improve the accuracy and effectiveness of various tasks, such as machine translation, image captioning, etc. Here, the benefit of attention layers in designing optical filters based on a stack of thin film materials is investigated. The superiority of Attention layers over fully-connected Deep Neural Networks is demonstrated for this task.
{"title":"On the Benefit of Attention in Inverse Design of Thin Films Filters","authors":"Barak Hadad, Omry Oren, A. Bahabad","doi":"10.1088/2632-2153/ad6832","DOIUrl":"https://doi.org/10.1088/2632-2153/ad6832","url":null,"abstract":"\u0000 Attention layers are a crucial component in many modern deep learning models, particularly those used in natural language processing and computer vision. Attention layers have been shown to improve the accuracy and effectiveness of various tasks, such as machine translation, image captioning, etc. Here, the benefit of attention layers in designing optical filters based on a stack of thin film materials is investigated. The superiority of Attention layers over fully-connected Deep Neural Networks is demonstrated for this task.","PeriodicalId":503691,"journal":{"name":"Machine Learning: Science and Technology","volume":"26 8","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141799450","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}
Pub Date : 2024-07-26DOI: 10.1088/2632-2153/ad6831
Delchere DON-TSA, Messanh Agbéko Mohou, K. Amouzouvi, Malik Maaza, K. Beltako
The high computational demand of the Density Functional Theory (DFT) based method for screening new materials properties remains a strong limitation to the development of clean and renewable energy technologies essential to transition to a carbon-neutral environment in the coming decades. Machine Learning comes into play with its innate capacity to handle huge amounts of data and high-dimensional statistical analysis. In this paper, supervised Machine Learning models together with data analysis on existing datasets obtained from a high-throughput calculation using Density Functional Theory are used to predict the Seebeck coefficient, electrical conductivity, and power factor of inorganic compounds. The analysis revealed a strong dependence of the thermoelectric properties on the effective masses, we also proposed a machine learning model for the prediction of highly performing thermoelectric materials which reached an efficiency of 95 percent. The analyzed data and developed model can significantly contribute to innovation by providing a faster and more accurate prediction of thermoelectric properties, thereby, facilitating the discovery of highly efficient thermoelectric materials.
{"title":"Predictive Models for Inorganic Materials Thermoelectric Properties with Machine Learning","authors":"Delchere DON-TSA, Messanh Agbéko Mohou, K. Amouzouvi, Malik Maaza, K. Beltako","doi":"10.1088/2632-2153/ad6831","DOIUrl":"https://doi.org/10.1088/2632-2153/ad6831","url":null,"abstract":"\u0000 The high computational demand of the Density Functional Theory (DFT) based method for screening new materials properties remains a strong limitation to the development of clean and renewable energy technologies essential to transition to a carbon-neutral environment in the coming decades. Machine Learning comes into play with its innate capacity to handle huge amounts of data and high-dimensional statistical analysis. In this paper, supervised Machine Learning models together with data analysis on existing datasets obtained from a high-throughput calculation using Density Functional Theory are used to predict the Seebeck coefficient, electrical conductivity, and power factor of inorganic compounds. The analysis revealed a strong dependence of the thermoelectric properties on the effective masses, we also proposed a machine learning model for the prediction of highly performing thermoelectric materials which reached an efficiency of 95 percent. The analyzed data and developed model can significantly contribute to innovation by providing a faster and more accurate prediction of thermoelectric properties, thereby, facilitating the discovery of highly efficient thermoelectric materials.","PeriodicalId":503691,"journal":{"name":"Machine Learning: Science and Technology","volume":"53 11","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141799973","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}
Pub Date : 2024-07-25DOI: 10.1088/2632-2153/ad674a
Sasaank Bandi, Chao Jiang, C. Marianetti
Machine learning approaches have recently emerged as powerful tools to probe structure-property relationships in crystals and molecules. Specifically, machine learning interatomic potentials (MLIPs) can accurately reproduce first-principles data at a cost similar to that of conventional interatomic potential approaches. While MLIPs have been extensively tested across various classes of materials and molecules, a clear characterization of the anharmonic terms encoded in the MLIPs is lacking. Here, we benchmark popular MLIPs using the anharmonic vibrational Hamiltonian of ThO2 in the fluorite crystal structure, which was constructed from density functional theory (DFT) using our highly accurate and efficient irreducible derivative methods. The anharmonic Hamiltonian was used to generate molecular dynamics (MD) trajectories, which were used to train three classes of MLIPs: Gaussian Approximation Potentials, Artificial Neural Networks (ANN), and Graph Neural Networks (GNN). The results were assessed by directly comparing phonons and their interactions, as well as phonon linewidths, phonon lineshifts, and thermal conductivity. The models were also trained on a DFT molecular dynamics dataset, demonstrating good agreement up to fifth-order for the ANN and GNN. Our analysis demonstrates that MLIPs have great potential for accurately characterizing anharmonicity in materials systems at a fraction of the cost of conventional first principles-based approaches.
{"title":"Benchmarking machine learning interatomic potentials via phonon anharmonicity","authors":"Sasaank Bandi, Chao Jiang, C. Marianetti","doi":"10.1088/2632-2153/ad674a","DOIUrl":"https://doi.org/10.1088/2632-2153/ad674a","url":null,"abstract":"\u0000 Machine learning approaches have recently emerged as powerful tools to probe structure-property relationships in crystals and molecules. Specifically, machine learning interatomic potentials (MLIPs) can accurately reproduce first-principles data at a cost similar to that of conventional interatomic potential approaches. While MLIPs have been extensively tested across various classes of materials and molecules, a clear characterization of the anharmonic terms encoded in the MLIPs is lacking. Here, we benchmark popular MLIPs using the anharmonic vibrational Hamiltonian of ThO2 in the fluorite crystal structure, which was constructed from density functional theory (DFT) using our highly accurate and efficient irreducible derivative methods. The anharmonic Hamiltonian was used to generate molecular dynamics (MD) trajectories, which were used to train three classes of MLIPs: Gaussian Approximation Potentials, Artificial Neural Networks (ANN), and Graph Neural Networks (GNN). The results were assessed by directly comparing phonons and their interactions, as well as phonon linewidths, phonon lineshifts, and thermal conductivity. The models were also trained on a DFT molecular dynamics dataset, demonstrating good agreement up to fifth-order for the ANN and GNN. Our analysis demonstrates that MLIPs have great potential for accurately characterizing anharmonicity in materials systems at a fraction of the cost of conventional first principles-based approaches.","PeriodicalId":503691,"journal":{"name":"Machine Learning: Science and Technology","volume":"23 15","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141803078","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}
Pub Date : 2024-07-25DOI: 10.1088/2632-2153/ad67a9
Cheng-Hong Yang, Tin Ho Cheung, Li-Yeh Chuang
Glioblastoma multiforme (GBM) is the most aggressive brain cancer in adults, with 3.2-3.4 cases per 100 thousand. In the US, brain cancer does not rank in the top 10 causes of death, but it remains in the top 15. Therefore, this research proposes a fuzzy-based GRUCoxPH model to identify missense variants associated with a high risk of all-cause mortality in GBM. The study combines various models, including fuzzy logic, Gated Recurrent Units (GRUs), and Cox Proportional Hazards Regression (CoxPh), to identify potential risk factors. The dataset is derived from TCGA-GBM clinicopathological information and mutations to create four risk score models: GRU, CoxPH, GRUCoxPHAddition, and GRUCoxPHMultiplication, analyzing 9 risk factors of the dataset. The Fuzzy-based GRUCoxPH model achieves an average accuracy of 86.97%, outperforming other models. This model demonstrates its ability to classify and identify missense variants associated with mortality in GBM, potentially advancing cancer research.
{"title":"Application of Deep Learning-based Fuzzy Systems to Analyze the Overall Risk of Mortality in Glioblastoma Multiforme","authors":"Cheng-Hong Yang, Tin Ho Cheung, Li-Yeh Chuang","doi":"10.1088/2632-2153/ad67a9","DOIUrl":"https://doi.org/10.1088/2632-2153/ad67a9","url":null,"abstract":"\u0000 Glioblastoma multiforme (GBM) is the most aggressive brain cancer in adults, with 3.2-3.4 cases per 100 thousand. In the US, brain cancer does not rank in the top 10 causes of death, but it remains in the top 15. Therefore, this research proposes a fuzzy-based GRUCoxPH model to identify missense variants associated with a high risk of all-cause mortality in GBM. The study combines various models, including fuzzy logic, Gated Recurrent Units (GRUs), and Cox Proportional Hazards Regression (CoxPh), to identify potential risk factors. The dataset is derived from TCGA-GBM clinicopathological information and mutations to create four risk score models: GRU, CoxPH, GRUCoxPHAddition, and GRUCoxPHMultiplication, analyzing 9 risk factors of the dataset. The Fuzzy-based GRUCoxPH model achieves an average accuracy of 86.97%, outperforming other models. This model demonstrates its ability to classify and identify missense variants associated with mortality in GBM, potentially advancing cancer research.","PeriodicalId":503691,"journal":{"name":"Machine Learning: Science and Technology","volume":"6 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141803423","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}
Pub Date : 2024-07-23DOI: 10.1088/2632-2153/ad66ae
A. Kesorn, Rutchapon Hunkao, Cheewawut Na Talang, Chanaprom Cholsuk, A. Sinsarp, Tobias Vogl, S. Suwanna, S. Yuma
We applied tree-based machine learning algorithms to predict the formation energy of impurities in 2D materials, where adsorbates and interstitial defects are investigated. Regression models based on random forest (RF), gradient boosting regression (GBR), histogram-based gradient-boosting regression (HGBR), and light gradient-boosting machine (LightGBM) algorithms are employed for training, testing, cross validation, and blind testing. We utilized chemical features from fundamental properties of atoms and supplemented them with structural features from the interaction of the added chemical element with its neighboring host atoms via the Jacobi-Legendre (JL) polynomials. Overall, the prediction accuracy yields optimal $text{MAE} approx 0.518$, $text{RMSE} approx 1.14$, and $R^2 approx 0.855$. When trained separately, we obtained lower residual errors RMSE and MAE, and higher $R^2$ value for predicting the formation energy in the adsorbates than in the interstitial defects. In both cases, the inclusion of the structural features via the JL polynomials improves the prediction accuracy of the formation energy in terms of decreasing RMSE and MAE, and increasing $R^2$. This work demonstrates the potential and application of physically meaningful features to obtain physical properties of impurities in 2D materials that otherwise would require higher computational cost.
{"title":"Formation Energy Prediction of Neutral Single-Atom Impurities in 2D Materials using Tree-based Machine Learning","authors":"A. Kesorn, Rutchapon Hunkao, Cheewawut Na Talang, Chanaprom Cholsuk, A. Sinsarp, Tobias Vogl, S. Suwanna, S. Yuma","doi":"10.1088/2632-2153/ad66ae","DOIUrl":"https://doi.org/10.1088/2632-2153/ad66ae","url":null,"abstract":"\u0000 We applied tree-based machine learning algorithms to predict the formation energy of impurities in 2D materials, where adsorbates and interstitial defects are investigated. Regression models based on random forest (RF), gradient boosting regression (GBR), histogram-based gradient-boosting regression (HGBR), and light gradient-boosting machine (LightGBM) algorithms are employed for training, testing, cross validation, and blind testing. We utilized chemical features from fundamental properties of atoms and supplemented them with structural features from the interaction of the added chemical element with its neighboring host atoms via the Jacobi-Legendre (JL) polynomials. Overall, the prediction accuracy yields optimal $text{MAE} approx 0.518$, $text{RMSE} approx 1.14$, and $R^2 approx 0.855$. When trained separately, we obtained lower residual errors RMSE and MAE, and higher $R^2$ value for predicting the formation energy in the adsorbates than in the interstitial defects. In both cases, the inclusion of the structural features via the JL polynomials improves the prediction accuracy of the formation energy in terms of decreasing RMSE and MAE, and increasing $R^2$. This work demonstrates the potential and application of physically meaningful features to obtain physical properties of impurities in 2D materials that otherwise would require higher computational cost.","PeriodicalId":503691,"journal":{"name":"Machine Learning: Science and Technology","volume":"118 51","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141811876","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}
Pub Date : 2024-07-17DOI: 10.1088/2632-2153/ad64a7
G. Galgóczi, Gábor Albrecht, G. Hamar, Dezső Varga
A machine learning algorithm (deep neural network) is presented to suppress background in muography applications mainly targeting volcanoes. Additionally it could be applied for large scale geological structures, such as ophiolites. The detector system investigated in this article is designed to suppress the low energy background by applying up to 5 lead absorber layers arranged among 8 detectors. This complicated system was simulated with a Monte-Carlo based particle simulation to provide teaching sample for the machine learning algorithm. It is shown that the developed deep neural network is capable of suppressing the low energy background considerably better than the classical tracking algorithm, therefore this additional suppression with machine learning yields in a significant improvement. The target areas of volcanoes lie beneath approximately a kilometer of rock that only fraction of a percent of muons have enough energy to penetrate. The machine learning algorithm takes advantage of the directional changes in the absorbers, as well as the correlation between the muons energy and the deposited energy in the detectors. Identifying very high energy muons is also a challenge: the classical algorithm discards considerable fraction of 1 TeV muons which create multiple hits due to brehmstrahlung, while the machine learning algorithm easily adapts to accept such patterns.
本文介绍了一种机器学习算法（深度神经网络），用于抑制主要针对火山的 muography 应用中的背景。此外，它还可应用于大型地质结构，如蛇绿岩。本文所研究的探测器系统是通过在 8 个探测器中应用多达 5 层铅吸收层来抑制低能量背景的。这个复杂的系统是用基于蒙特卡洛粒子模拟的方法模拟的，为机器学习算法提供了教学样本。结果表明，所开发的深度神经网络能够比经典跟踪算法更好地抑制低能量背景，因此，这种额外的机器学习抑制效果显著提高。火山的目标区域位于大约一千米的岩石之下，只有百分之一的μ介子有足够的能量穿透岩石。机器学习算法利用了吸收器的方向变化以及μ介子能量与探测器沉积能量之间的相关性。识别高能μ介子也是一项挑战：经典算法会丢弃相当一部分 1 TeV μ介子，这些μ介子会因轫致辐射而产生多次撞击，而机器学习算法很容易适应这种模式。
{"title":"Background suppression for volcano muography with machine learning","authors":"G. Galgóczi, Gábor Albrecht, G. Hamar, Dezső Varga","doi":"10.1088/2632-2153/ad64a7","DOIUrl":"https://doi.org/10.1088/2632-2153/ad64a7","url":null,"abstract":"\u0000 A machine learning algorithm (deep neural network) is presented to suppress background in muography applications mainly targeting volcanoes. Additionally it could be applied for large scale geological structures, such as ophiolites. The detector system investigated in this article is designed to suppress the low energy background by applying up to 5 lead absorber layers arranged among 8 detectors. This complicated system was simulated with a Monte-Carlo based particle simulation to provide teaching sample for the machine learning algorithm. It is shown that the developed deep neural network is capable of suppressing the low energy background considerably better than the classical tracking algorithm, therefore this additional suppression with machine learning yields in a significant improvement. The target areas of volcanoes lie beneath approximately a kilometer of rock that only fraction of a percent of muons have enough energy to penetrate. The machine learning algorithm takes advantage of the directional changes in the absorbers, as well as the correlation between the muons energy and the deposited energy in the detectors. Identifying very high energy muons is also a challenge: the classical algorithm discards considerable fraction of 1 TeV muons which create multiple hits due to brehmstrahlung, while the machine learning algorithm easily adapts to accept such patterns.","PeriodicalId":503691,"journal":{"name":"Machine Learning: Science and Technology","volume":" 47","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141830240","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}
Pub Date : 2024-07-16DOI: 10.1088/2632-2153/ad63f3
Georgia Papacharalampous, Hristos Tyralis, N. Doulamis, Anastasios Doulamis
Merging satellite and gauge data with machine learning produces high-resolution precipitation datasets, but uncertainty estimates are often missing. We addressed the gap of how to optimally provide such estimates by benchmarking six algorithms, mostly novel even for the more general task of quantifying predictive uncertainty in spatial prediction settings. On 15 years of monthly data from over the contiguous United States (CONUS), we compared quantile regression (QR), quantile regression forests (QRF), generalized random forests (GRF), gradient boosting machines (GBM), light gradient boosting machines (LightGBM), and quantile regression neural networks (QRNN). Their ability to issue predictive precipitation quantiles at nine quantile levels (0.025, 0.050, 0.100, 0.250, 0.500, 0.750, 0.900, 0.950, 0.975), approximating the full probability distribution, was evaluated using quantile scoring functions and the quantile scoring rule. Predictors at a site were nearby values from two satellite precipitation retrievals, namely PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) and IMERG (Integrated Multi-satellitE Retrievals), and the site’s elevation. The dependent variable was the monthly mean gauge precipitation. With respect to QR, LightGBM showed improved performance in terms of the quantile scoring rule by 11.10%, also surpassing QRF (7.96%), GRF (7.44%), GBM (4.64%) and QRNN (1.73%). Notably, LightGBM outperformed all random forest variants, the current standard in spatial prediction with machine learning. To conclude, we propose a suite of machine learning algorithms for estimating uncertainty in spatial data prediction, supported with a formal evaluation framework based on scoring functions and scoring rules.
{"title":"Uncertainty estimation of machine learning spatial precipitation predictions from satellite data","authors":"Georgia Papacharalampous, Hristos Tyralis, N. Doulamis, Anastasios Doulamis","doi":"10.1088/2632-2153/ad63f3","DOIUrl":"https://doi.org/10.1088/2632-2153/ad63f3","url":null,"abstract":"\u0000 Merging satellite and gauge data with machine learning produces high-resolution precipitation datasets, but uncertainty estimates are often missing. We addressed the gap of how to optimally provide such estimates by benchmarking six algorithms, mostly novel even for the more general task of quantifying predictive uncertainty in spatial prediction settings. On 15 years of monthly data from over the contiguous United States (CONUS), we compared quantile regression (QR), quantile regression forests (QRF), generalized random forests (GRF), gradient boosting machines (GBM), light gradient boosting machines (LightGBM), and quantile regression neural networks (QRNN). Their ability to issue predictive precipitation quantiles at nine quantile levels (0.025, 0.050, 0.100, 0.250, 0.500, 0.750, 0.900, 0.950, 0.975), approximating the full probability distribution, was evaluated using quantile scoring functions and the quantile scoring rule. Predictors at a site were nearby values from two satellite precipitation retrievals, namely PERSIANN (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) and IMERG (Integrated Multi-satellitE Retrievals), and the site’s elevation. The dependent variable was the monthly mean gauge precipitation. With respect to QR, LightGBM showed improved performance in terms of the quantile scoring rule by 11.10%, also surpassing QRF (7.96%), GRF (7.44%), GBM (4.64%) and QRNN (1.73%). Notably, LightGBM outperformed all random forest variants, the current standard in spatial prediction with machine learning. To conclude, we propose a suite of machine learning algorithms for estimating uncertainty in spatial data prediction, supported with a formal evaluation framework based on scoring functions and scoring rules.","PeriodicalId":503691,"journal":{"name":"Machine Learning: Science and Technology","volume":"7 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141641258","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}
This study explores the implementation of Physics-Informed Neural Networks (PINN) to analyze turbulent flow in composite porous-fluid systems. These systems are composed of a fluid-saturated porous medium and an adjacent fluid, where the flow properties are exchanged across the porous-fluid interface. The PINN model employs a novel approach combining supervised learning and enforces fidelity to flow physics through penalization by the Reynolds-Averaged Navier-Stokes (RANS) equations. Two cases were simulated for this purpose: solid block, i.e., porous media with zero porosity, and porous block with a defined porosity. The effect of providing internal training data on the accuracy of the PINN predictions for prominent flow features including leakage, channeling effect and wake recirculation were investigated. Additionally, L2 norm error, which evaluates the prediction accuracy for flow variables was studied. Furthermore, PINN training time in both cases with internal training data were considered in this study. The results showed that the PINN predictions achieved high accuracy for the prominent flow features compared to the reference RANS data. In addition, second-order internal training data in the wall-normal direction reduced the L2 norm error by 100% for the solid block case, while for the porous block case, providing training data at the porous-fluid interface, increased the prediction accuracy by nearly 40% for second-order statistics. The study elucidates the impact of the internal training data distribution on the PINN training in complex turbulent flow dynamics, underscoring the necessity of turbulent second-order statistics variables in PINN training and an additional velocity gradient treatment to enhance PINN prediction.
{"title":"Physics-Informed Neural Network for Turbulent Flow Reconstruction in Composite Porous-Fluid Systems","authors":"Seohee Jang, Mohammad Jadidi, Saleh Rezaeiravesh, Alistair Revell, Yasser Mahmoudi Larimi","doi":"10.1088/2632-2153/ad63f4","DOIUrl":"https://doi.org/10.1088/2632-2153/ad63f4","url":null,"abstract":"\u0000 This study explores the implementation of Physics-Informed Neural Networks (PINN) to analyze turbulent flow in composite porous-fluid systems. These systems are composed of a fluid-saturated porous medium and an adjacent fluid, where the flow properties are exchanged across the porous-fluid interface. The PINN model employs a novel approach combining supervised learning and enforces fidelity to flow physics through penalization by the Reynolds-Averaged Navier-Stokes (RANS) equations. Two cases were simulated for this purpose: solid block, i.e., porous media with zero porosity, and porous block with a defined porosity. The effect of providing internal training data on the accuracy of the PINN predictions for prominent flow features including leakage, channeling effect and wake recirculation were investigated. Additionally, L2 norm error, which evaluates the prediction accuracy for flow variables was studied. Furthermore, PINN training time in both cases with internal training data were considered in this study. The results showed that the PINN predictions achieved high accuracy for the prominent flow features compared to the reference RANS data. In addition, second-order internal training data in the wall-normal direction reduced the L2 norm error by 100% for the solid block case, while for the porous block case, providing training data at the porous-fluid interface, increased the prediction accuracy by nearly 40% for second-order statistics. The study elucidates the impact of the internal training data distribution on the PINN training in complex turbulent flow dynamics, underscoring the necessity of turbulent second-order statistics variables in PINN training and an additional velocity gradient treatment to enhance PINN prediction.","PeriodicalId":503691,"journal":{"name":"Machine Learning: Science and Technology","volume":"4 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141641606","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}
Pub Date : 2024-07-16DOI: 10.1088/2632-2153/ad5fdc
M. Alshayeji
Machine learning techniques applied in neuroimaging have prompted researchers to build models for early diagnosis of brain illnesses such as Alzheimer’s disease (AD). Although this task is difficult, advanced deep-learning (DL) approaches can be used. These DL models are effective, but difficult to interpret, time-consuming, and resource-intensive. Therefore, neuroscientists are interested in employing novel, less complex structures such as transformers that have superior pattern-extraction capabilities. In this study, an automated framework for accurate AD diagnosis and precise stage identification was developed by employing vision transformers (ViTs) with fewer computational resources. ViT, which captures the global context as opposed to convolutional neural networks (CNNs) with local receptive fields, is more efficient for brain image processing than CNN because the brain is a highly complex network with connected parts. The self-attention mechanism in the ViT helps to achieve this goal. Magnetic resonance brain images belonging to four stages were utilized to develop the proposed model, which achieved 99.83% detection accuracy, 99.69% sensitivity, 99.88% specificity, and 0.17% misclassification rate. Moreover, to prove the ability of the model to generalize, the mean distances of the transformer blocks and attention heat maps were visualized to understand what the model learned from the MRI input image.
{"title":"Alzheimer’s disease detection and stage identification from magnetic resonance brain images using vision transformer","authors":"M. Alshayeji","doi":"10.1088/2632-2153/ad5fdc","DOIUrl":"https://doi.org/10.1088/2632-2153/ad5fdc","url":null,"abstract":"\u0000 Machine learning techniques applied in neuroimaging have prompted researchers to build models for early diagnosis of brain illnesses such as Alzheimer’s disease (AD). Although this task is difficult, advanced deep-learning (DL) approaches can be used. These DL models are effective, but difficult to interpret, time-consuming, and resource-intensive. Therefore, neuroscientists are interested in employing novel, less complex structures such as transformers that have superior pattern-extraction capabilities. In this study, an automated framework for accurate AD diagnosis and precise stage identification was developed by employing vision transformers (ViTs) with fewer computational resources. ViT, which captures the global context as opposed to convolutional neural networks (CNNs) with local receptive fields, is more efficient for brain image processing than CNN because the brain is a highly complex network with connected parts. The self-attention mechanism in the ViT helps to achieve this goal. Magnetic resonance brain images belonging to four stages were utilized to develop the proposed model, which achieved 99.83% detection accuracy, 99.69% sensitivity, 99.88% specificity, and 0.17% misclassification rate. Moreover, to prove the ability of the model to generalize, the mean distances of the transformer blocks and attention heat maps were visualized to understand what the model learned from the MRI input image.","PeriodicalId":503691,"journal":{"name":"Machine Learning: Science and Technology","volume":"1 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141641279","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}
Pub Date : 2024-07-15DOI: 10.1088/2632-2153/ad6391
Naoki Koyama, Yusuke Sakai, Seiya Sasaoka, Diego Dominguez, K. Somiya, Yuto Omae, Yoshikazu Terada, M. Meyer-Conde, Hirotaka Takahashi
In the pursuit of detecting gravitational waves, ground-based interferometers (e.g. LIGO, Virgo, and KAGRA) face a significant challenge: achieving the extremely high sensitivity required to detect fluctuations at distances significantly smaller than the diameter of an atomic nucleus. Cutting-edge materials and innovative engineering techniques have been employed to enhance the stability and precision of the interferometer apparatus over the years. These efforts are crucial for reducing the noise that masks the subtle gravitational wave signals. Various sources of interference, such as seismic activity, thermal fluctuations, and other environmental factors, contribute to the total noise spectra characteristic of the detector. Therefore, addressing these sources is essential to enhance the interferometer apparatus's stability and precision. Recent research has emphasised the importance of classifying non-stationary and non-Gaussian glitches, employing sophisticated algorithms and machine learning methods to distinguish genuine gravitational wave signals from instrumental artefacts. The time-frequency-amplitude representation of these transient disturbances exhibits a wide range of new shapes, variability, and features, reflecting the evolution of interferometer technology. In this study, we developed a convolutional neural network model to classify glitches using spectrogram images from the Gravity Spy O1 dataset. We employed score-class activation mapping and the uniform manifold approximation and projection algorithm to visualise and understand the classification decisions made by our model. We assessed the model's validity and investigated the causes of misclassification from these results.
{"title":"Enhancing the Rationale of Convolutional Neural Networks for Glitch Classification in Gravitational Wave Detectors: A Visual Explanation","authors":"Naoki Koyama, Yusuke Sakai, Seiya Sasaoka, Diego Dominguez, K. Somiya, Yuto Omae, Yoshikazu Terada, M. Meyer-Conde, Hirotaka Takahashi","doi":"10.1088/2632-2153/ad6391","DOIUrl":"https://doi.org/10.1088/2632-2153/ad6391","url":null,"abstract":"\u0000 In the pursuit of detecting gravitational waves, ground-based interferometers (e.g. LIGO, Virgo, and KAGRA) face a significant challenge: achieving the extremely high sensitivity required to detect fluctuations at distances significantly smaller than the diameter of an atomic nucleus. Cutting-edge materials and innovative engineering techniques have been employed to enhance the stability and precision of the interferometer apparatus over the years. These efforts are crucial for reducing the noise that masks the subtle gravitational wave signals. Various sources of interference, such as seismic activity, thermal fluctuations, and other environmental factors, contribute to the total noise spectra characteristic of the detector. Therefore, addressing these sources is essential to enhance the interferometer apparatus's stability and precision. Recent research has emphasised the importance of classifying non-stationary and non-Gaussian glitches, employing sophisticated algorithms and machine learning methods to distinguish genuine gravitational wave signals from instrumental artefacts. The time-frequency-amplitude representation of these transient disturbances exhibits a wide range of new shapes, variability, and features, reflecting the evolution of interferometer technology. In this study, we developed a convolutional neural network model to classify glitches using spectrogram images from the Gravity Spy O1 dataset. We employed score-class activation mapping and the uniform manifold approximation and projection algorithm to visualise and understand the classification decisions made by our model. We assessed the model's validity and investigated the causes of misclassification from these results.","PeriodicalId":503691,"journal":{"name":"Machine Learning: Science and Technology","volume":"9 9","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141645927","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：