scholarly journals Machine Learning for Fluid Mechanics

2020 ◽  
Vol 52 (1) ◽  
pp. 477-508 ◽  
Author(s):  
Steven L. Brunton ◽  
Bernd R. Noack ◽  
Petros Koumoutsakos
Keyword(s):  
Machine Learning ◽  
Fluid Mechanics ◽  
Domain Knowledge ◽  
Large Scale ◽  
Past History ◽  
Fluid Flows ◽  
Field Measurements ◽  
Industrial Applications ◽  
Current Lines ◽  
Large Scale Simulations

The field of fluid mechanics is rapidly advancing, driven by unprecedented volumes of data from experiments, field measurements, and large-scale simulations at multiple spatiotemporal scales. Machine learning (ML) offers a wealth of techniques to extract information from data that can be translated into knowledge about the underlying fluid mechanics. Moreover, ML algorithms can augment domain knowledge and automate tasks related to flow control and optimization. This article presents an overview of past history, current developments, and emerging opportunities of ML for fluid mechanics. We outline fundamental ML methodologies and discuss their uses for understanding, modeling, optimizing, and controlling fluid flows. The strengths and limitations of these methods are addressed from the perspective of scientific inquiry that considers data as an inherent part of modeling, experiments, and simulations. ML provides a powerful information-processing framework that can augment, and possibly even transform, current lines of fluid mechanics research and industrial applications.

scholarly journals A Machine Learning Approach for the Discovery of Ligand-Specific Functional Mechanisms of GPCRs

Molecules ◽  
2019 ◽  
Vol 24 (11) ◽  
pp. 2097 ◽  
Author(s):  
Ambrose Plante ◽  
Derek M. Shore ◽  
Giulia Morra ◽  
George Khelashvili ◽  
Harel Weinstein
Keyword(s):  
Machine Learning ◽  
Big Data ◽  
Large Scale ◽  
Molecular Mechanisms ◽  
Md Simulations ◽  
Machine Learning Approach ◽  
And Function ◽  
Functional Mechanisms ◽  
G Protein Coupled ◽  
Large Scale Simulations

G protein-coupled receptors (GPCRs) play a key role in many cellular signaling mechanisms, and must select among multiple coupling possibilities in a ligand-specific manner in order to carry out a myriad of functions in diverse cellular contexts. Much has been learned about the molecular mechanisms of ligand-GPCR complexes from Molecular Dynamics (MD) simulations. However, to explore ligand-specific differences in the response of a GPCR to diverse ligands, as is required to understand ligand bias and functional selectivity, necessitates creating very large amounts of data from the needed large-scale simulations. This becomes a Big Data problem for the high dimensionality analysis of the accumulated trajectories. Here we describe a new machine learning (ML) approach to the problem that is based on transforming the analysis of GPCR function-related, ligand-specific differences encoded in the MD simulation trajectories into a representation recognizable by state-of-the-art deep learning object recognition technology. We illustrate this method by applying it to recognize the pharmacological classification of ligands bound to the 5-HT2A and D2 subtypes of class-A GPCRs from the serotonin and dopamine families. The ML-based approach is shown to perform the classification task with high accuracy, and we identify the molecular determinants of the classifications in the context of GPCR structure and function. This study builds a framework for the efficient computational analysis of MD Big Data collected for the purpose of understanding ligand-specific GPCR activity.

scholarly journals Integrate multi-omics data with biological interaction networks using Multi-view Factorization AutoEncoder (MAE)

BMC Genomics ◽  
2019 ◽  
Vol 20 (S11) ◽  
Author(s):  
Tianle Ma ◽  
Aidong Zhang
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Domain Knowledge ◽  
Large Scale ◽  
Learning Model ◽  
Interaction Networks ◽  
Omics Data ◽  
Biological Interaction ◽  
Molecular Features ◽  
Training Objective

Abstract Background Comprehensive molecular profiling of various cancers and other diseases has generated vast amounts of multi-omics data. Each type of -omics data corresponds to one feature space, such as gene expression, miRNA expression, DNA methylation, etc. Integrating multi-omics data can link different layers of molecular feature spaces and is crucial to elucidate molecular pathways underlying various diseases. Machine learning approaches to mining multi-omics data hold great promises in uncovering intricate relationships among molecular features. However, due to the “big p, small n” problem (i.e., small sample sizes with high-dimensional features), training a large-scale generalizable deep learning model with multi-omics data alone is very challenging. Results We developed a method called Multi-view Factorization AutoEncoder (MAE) with network constraints that can seamlessly integrate multi-omics data and domain knowledge such as molecular interaction networks. Our method learns feature and patient embeddings simultaneously with deep representation learning. Both feature representations and patient representations are subject to certain constraints specified as regularization terms in the training objective. By incorporating domain knowledge into the training objective, we implicitly introduced a good inductive bias into the machine learning model, which helps improve model generalizability. We performed extensive experiments on the TCGA datasets and demonstrated the power of integrating multi-omics data and biological interaction networks using our proposed method for predicting target clinical variables. Conclusions To alleviate the overfitting problem in deep learning on multi-omics data with the “big p, small n” problem, it is helpful to incorporate biological domain knowledge into the model as inductive biases. It is very promising to design machine learning models that facilitate the seamless integration of large-scale multi-omics data and biomedical domain knowledge for uncovering intricate relationships among molecular features and clinical features.

Complex Flow Dynamics in Dense Granular Flows—Part II: Simulations

2006 ◽  
Vol 74 (4) ◽  
pp. 691-702 ◽  
Author(s):  
Piroz Zamankhan ◽  
Jun Huang
Keyword(s):  
Granular Flow ◽  
Large Scale ◽  
Flow Behavior ◽  
Granular Flows ◽  
Industrial Applications ◽  
Complex Flow ◽  
Flow Phenomena ◽  
Newton’S Cradle ◽  
Large Scale Simulations ◽  
Dense Granular Flows

By applying a methodology useful for analysis of complex fluids based on a synergistic combination of experiments, computer simulations, and theoretical investigation, a model was built to investigate the fluid dynamics of granular flows in an intermediate regime, where both collisional and frictional interactions may affect the flow behavior. In Part I, experiments were described using a modified Newton’s Cradle device to obtain values for the viscous damping coefficient, which were scarce in the literature. This paper discusses detailed simulations of frictional interactions between the grains during a binary collision by employing a numerical model based on finite element methods. Numerical results are presented of slipping, and sticking motions of a first grain over the second one. The key was to utilize the results of the aforementioned comprehensive model in order to provide a simplified model for accurate and efficient granular-flow simulations with which the qualitative trends observed in the experiments can be captured. To validate the model, large scale simulations were performed for the specific case of granular flow in a rapidly spinning bucket. The model was able to reproduce experimentally observed flow phenomena, such as the formation of a depression in the center of the bucket spinning at high frequency of 100rad/s. This agreement suggests that the model may be a useful tool for the prediction of dense granular flows in industrial applications, but highlights the need for further experimental investigation of granular flows in order to refine the model.

scholarly journals Optimal Planning Method for Large-Scale Historical Exhibits in the Taiwan Railway Museum

2021 ◽  
Vol 11 (5) ◽  
pp. 2424
Author(s):  
Lin Pey Fan ◽  
Tzu How Chu
Keyword(s):  
Machine Learning ◽  
Domain Knowledge ◽  
Large Scale ◽  
Low Cost ◽  
Planning Method ◽  
Optimal Planning ◽  
Innovative Strategy ◽  
Heritage Sites ◽  
Progressive Development ◽  
Point Line

The curation design of cultural heritage sites, such as museums, influence the level of visitor satisfaction and the possibility of revisitation; therefore, an efficient exhibit layout is critical. The difficulty of determining the behavior of visitors and the layout of galleries means that exhibition layout is a knowledge-intensive, time-consuming process. The progressive development of machine learning provides a low-cost and highly flexible workflow in the management of museums, compared to traditional curation design. For example, the facility’s optimal layout, floor, and furniture arrangement can be obtained through the repeated adjustment of artificial intelligence algorithms within a relatively short time. In particular, an optimal planning method is indispensable for the immense and heavy trains in the railway museum. In this study, we created an innovative strategy to integrate the domain knowledge of exhibit displaying, spatial planning, and machine learning to establish a customized recommendation scheme. Guided by an interactive experience model and the morphology of point–line–plane–stereo, we obtained three aspects (visitors, objects, and space), 12 dimensions (orientation, visiting time, visual distance, centrality, main path, district, capacity, etc.), 30 physical principles, 24 suggestions, and five main procedures to implement layout patterns and templates to create an exhibit layout guide for the National Railway Museum of Taiwan, which is currently being transferred from the railway workshop for the sake of preserving the rail culture heritage. Our results are suitable and extendible to different museums by adjusting the criteria used to establish a new recommendation scheme.

scholarly journals Automated Ontology Building for News Text using Associative Word Properties

2020 ◽  
Vol 9 (5) ◽  
pp. 663-668
Keyword(s):  
Machine Learning ◽  
Domain Knowledge ◽  
Large Scale ◽  
Performance Enhancement ◽  
Text Retrieval ◽  
Retrieval Performance ◽  
New Words ◽  
Dynamic Events ◽  
Ontology Building

It is known in the literature that ontology had been used extensively in machine learning for performance enhancement for text retrieval. It is also shown that robust ontology with detailed description of the domain knowledge will contribute to the accuracy in the retrieval. Nevertheless, we argue in some domain such as news text retrieval, building an ontology manually can be costly for a large-scale news repository and especially with the changes in content due to the dynamic events. In addition, maintenance can be a dauting task to keep up with new words that are associated with new events. This paper demonstrates the attempt to fully automate the development of an ontology for identifying the news domain and its subdomain. The ontology specification is defined based on the needs of the accuracy in retrieval. The mechanism of generating the ontology specification is defined and the results of the retrieval performance is discussed.

scholarly journals A Machine Learning Approach for the Discovery of Ligand-specific Functional Mechanisms of GPCRs

2019 ◽  
Author(s):  
Ambrose Plante ◽  
Derek M. Shore ◽  
Giulia Morra ◽  
George Khelashvili ◽  
Harel Weinstein
Keyword(s):  
Machine Learning ◽  
Big Data ◽  
Large Scale ◽  
Molecular Mechanisms ◽  
Md Simulations ◽  
Machine Learning Approach ◽  
And Function ◽  
Functional Mechanisms ◽  
G Protein Coupled ◽  
Large Scale Simulations

G protein-coupled receptors (GPCRs) play a key role in many cellular signaling mechanisms, and must select among multiple coupling possibilities in a ligand-specific manner in order to carry out a myriad of functions in diverse cellular contexts. Much has been learned about the molecular mechanisms of ligand-GPCR complexes from Molecular Dynamics (MD) simulations. However, to explore ligand-specific differences in the response of a GPCR to diverse ligands, as is required to understand ligand bias and functional selectivity, necessitates creating very large amounts of data from the needed large-scale simulations. This becomes a Big Data problem for the high dimensionality analysis of the accumulated trajectories. Here we describe a new machine learning (ML) approach to the problem that is based on transforming the analysis of GPCR function-related, ligand-specific differences encoded in the MD simulation trajectories into a representation recognizable by state-of-the-art deep learning object recognition technology. We illustrate this method by applying it to recognize the pharmacological classification of ligands bound to the 5-HT2A and D2 subtypes of class A GPCRs from the serotonin and dopamine families. The ML-based approach is shown to perform the classification task with high accuracy, and we identify the molecular determinants of the classifications in the context of GPCR structure and function. This study builds a framework for the efficient computational analysis of MD Big Data collected for the purpose of understanding ligand-specific GPCR activity.

scholarly journals Semantic Data Mining in Ubiquitous Sensing: A Survey

Sensors ◽  
2021 ◽  
Vol 21 (13) ◽  
pp. 4322
Author(s):  
Grzegorz J. Nalepa ◽  
Szymon Bobek ◽  
Krzysztof Kutt ◽  
Martin Atzmueller
Keyword(s):  
Data Mining ◽  
Domain Knowledge ◽  
Large Scale ◽  
Industrial Applications ◽  
Levels Of Abstraction ◽  
Current Application ◽  
Collective Behaviors ◽  
Semantic Data ◽  
Ubiquitous Sensing ◽  
Semantic Data Mining

Mining ubiquitous sensing data is important but also challenging, due to many factors, such as heterogeneous large-scale data that is often at various levels of abstraction. This also relates particularly to the important aspects of the explainability and interpretability of the applied models and their results, and thus ultimately to the outcome of the data mining process. With this, in general, the inclusion of domain knowledge leading towards semantic data mining approaches is an emerging and important research direction. This article aims to survey relevant works in these areas, focusing on semantic data mining approaches and methods, but also on selected applications of ubiquitous sensing in some of the most prominent current application areas. Here, we consider in particular: (1) environmental sensing; (2) ubiquitous sensing in industrial applications of artificial intelligence; and (3) social sensing relating to human interactions and the respective individual and collective behaviors. We discuss these in detail and conclude with a summary of this emerging field of research. In addition, we provide an outlook on future directions for semantic data mining in ubiquitous sensing contexts.

Sign in / Sign up

Export Citation Format

Share Document