Enabling large‐scale viscoelastic calculations via neural network acceleration

DeVries, Phoebe M. R.; Thompson, T. B.; Meade, Brendan J.

doi:10.1002/2017gl072716

Cited by 54 publications

(27 citation statements)

References 29 publications

(35 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous models took hundreds of hours to perform these calculations. Deep neural network is used in this experiment, and the results are astonishing; DNN speeds up the system by more than 50,000% (DeVries et al, ).…”

Section: Applications Of Deep Learningmentioning

confidence: 99%

See 1 more Smart Citation

A systematic review on deep learning architectures and applications

Khamparia

Singh

2019

Expert Systems

View full text Add to dashboard Cite

The amount of digital data in the universe is growing at an exponential rate, doubling every 2 years, and changing how we live in the world. The information storage capacity and data requirement crossed the zettabytes. With this level of bombardment of data on machine learning techniques, it becomes very difficult to carry out parallel computations. Deep learning is broadening its scope and gaining more popularity in natural language processing, feature extraction and visualization, and almost in every machine learning trend. The purpose of this study is to provide a brief review of deep learning architectures and their working. Research papers and proceedings of conferences from various authentic resources (Institute of Electrical and Electronics Engineers, Wiley, Nature, and Elsevier) are studied and analyzed. Different architectures and their effectiveness to solve domain specific problems are evaluated. Various limitations and open problems of current architectures are discussed to provide better insights to help researchers and student to resume their research on these issues. One hundred one articles were reviewed for this meta‐analysis of deep learning. From this analysis, it is concluded that advanced deep learning architectures are combinations of few conventional architectures. For example, deep belief network and convolutional neural network are used to build convolutional deep belief network, which has higher capabilities than the parent architectures. These combined architectures are more robust to explore the problem space and thus can be the answer to build a general‐purpose architecture.

show abstract

Section: Applications Of Deep Learningmentioning

confidence: 99%

“…Convolutional neural network has used over an image, and RNN helps to learn the description of an image. Figure 12 shows an example of the output of this model (DeVries, Thompson, & Meade, 2017).…”

Section: Image Description or Caption Generationmentioning

confidence: 99%

A systematic review on deep learning architectures and applications

Khamparia

Singh

2019

Expert Systems

View full text Add to dashboard Cite

show abstract

“…However, as our abilities in obtaining more reliable data at higher resolution both in space and time increase, our demands for alternate techniques for processing and manipulating large amount of data also increase. Although the application of machine learning techniques in some scientific and nonscientific fields extends to few decades ago (e.g., Dysart, 1996), they have been employed increasingly in the past years for classification and event predictions, for example, in geoscience (DeVries et al, 2017), biology (Hunter, 1990;Rawlings & Fox, 1994;Vidyasagar, 2014), medicine (Mena et al, 2012), medical diagnosis and prediction (Chen et al, 2007;Foster et al, 2014), astrophysics (Banerji et al, 2010;Polsterer et al, 2014;VanderPlas et al, 2012), and even in fusion power (Tang et al, 2016). Atkins et al (2016) employ Bayesian learning method to constrain selected mantle parameters.…”

Section: Introductionmentioning

confidence: 99%

“…Atkins et al (2016) employ Bayesian learning method to constrain selected mantle parameters. Although the application of machine learning techniques in some scientific and nonscientific fields extends to few decades ago (e.g., Dysart, 1996), they have been employed increasingly in the past years for classification and event predictions, for example, in geoscience (DeVries et al, 2017), biology (Hunter, 1990;Rawlings & Fox, 1994;Vidyasagar, 2014), medicine (Mena et al, 2012), medical diagnosis and prediction (Chen et al, 2007;Foster et al, 2014), astrophysics (Banerji et al, 2010;Polsterer et al, 2014;VanderPlas et al, 2012), and even in fusion power (Tang et al, 2016). In recent years machine learning techniques have found their application in many fields of geoscience and the related fields as groundwater studies (Paradis et al, 2014;Sahoo et al, 2017), predictions of earthquakes from laboratory experiments (Rouet-Leduc et al, 2017), prediction of seafloor sediment porosity (Martin et al, 2015), forecasting of ozone peaks and concentration based on the previous observations (Mallet et al, 2009), SHAHNAS ET AL.…”

Section: Introductionmentioning

confidence: 99%

Inverse Problems in Geodynamics Using Machine Learning Algorithms

2018

View full text Add to dashboard Cite

During the past few decades numerical studies have been widely employed to explore the style of circulation and mixing in the mantle of Earth and other planets. However, in geodynamical studies there are many properties from mineral physics, geochemistry, and petrology in these numerical models. Machine learning, as a computational statistic‐related technique and a subfield of artificial intelligence, has rapidly emerged recently in many fields of sciences and engineering. We focus here on the application of supervised machine learning (SML) algorithms in predictions of mantle flow processes. Specifically, we emphasize on estimating mantle properties by employing machine learning techniques in solving an inverse problem. Using snapshots of numerical convection models as training samples, we enable machine learning models to determine the magnitude of the spin transition‐induced density anomalies that can cause flow stagnation at midmantle depths. Employing support vector machine algorithms, we show that SML techniques can successfully predict the magnitude of mantle density anomalies and can also be used in characterizing mantle flow patterns. The technique can be extended to more complex geodynamic problems in mantle dynamics by employing deep learning algorithms for putting constraints on properties such as viscosity, elastic parameters, and the nature of thermal and chemical anomalies.

show abstract

“…Computational expense is often a major limitation of real-time forecasting systems [9,10]. Here, we apply machine learning techniques to predict wave conditions with the goal of replacing a computationally intensive physics-based model by straightforward multiplication of an input vector by mapping matrices resulting from the trained machine learning models.…”

mentioning

confidence: 99%

A machine learning framework to forecast wave conditions

James

Zhang

O’Donncha

2018

Coastal Engineering

273

View full text Add to dashboard Cite

A machine learning framework is developed to estimate ocean-wave conditions. By supervised training of machine learning models on many thousands of iterations of a physics-based wave model, accurate representations of significant wave heights and period can be used to predict ocean conditions. A model of Monterey Bay was used as the example test site; it was forced by measured wave conditions, ocean-current nowcasts, and reported winds.These input data along with model outputs of spatially variable wave heights and characteristic period were aggregated into supervised learning training and test data sets, which were supplied to machine learning models. These machine learning models replicated wave heights with a root-mean-squared error of 9 cm and correctly identify over 90% of the characteristic periods for the test-data sets. Impressively, transforming model inputs to outputs through matrix operations requires only a fraction (< 1/1, 000 th ) of the computation time compared to forecasting with the physics-based model. There are myriad reasons why predicting wave conditions is important to the economy. Surfers aside, there are fundamental reasons why knowledge of wave conditions for the next couple of days is important. For example, shipping routes can be optimized by avoiding rough seas thereby reducing shipping times. Another industry that benefits from knowledge of wave conditions is the $160B (2014) aquaculture industry [1], which could optimize harvesting operations accordingly. Knowledge of littoral conditions is critical to military and amphibious operations by Navy and Marine Corps teams. Also, predicting the energy production from renewable energy sources is critical to maintaining a stable electrical grid because many renewable energy sources (e.g., solar, wind, tidal, wave, etc.) are intermittent. For deeper market penetration of renewable energies, combinations of increased energy storage and improved energy-generation predictions will be required. The US Department of Energy has recently invested in the design, permitting, and construction of an open-water, grid-connected national Wave Energy Test Facility at Oregon State University [2]. Given that America's technically recoverable wave-energy resource is up to 1,230 TW-hr [3], there is a strong interest in developing this renewable resource [4]. Commercializationand deployment of wave-energy technologies will require not only addressing permitting and regulatory matters, but overcoming technological challenges, one of which is being able to provide an accurate prediction of energy generation. A requirement for any forecast is that an appropriately representative model be developed, calibrated, and validated. Moreover, this model must be

show abstract

Enabling large‐scale viscoelastic calculations via neural network acceleration

Cited by 54 publications

References 29 publications

A systematic review on deep learning architectures and applications

A systematic review on deep learning architectures and applications

Inverse Problems in Geodynamics Using Machine Learning Algorithms

A machine learning framework to forecast wave conditions

Contact Info

Product

Resources

About