Deep neural networks are gaining importance and popularity in applications and services.Due to the enormous number of learnable parameters and datasets,the training of neural networks is computationally costly.Paralle...Deep neural networks are gaining importance and popularity in applications and services.Due to the enormous number of learnable parameters and datasets,the training of neural networks is computationally costly.Parallel and distributed computation-based strategies are used to accelerate this training process.Generative Adversarial Networks(GAN)are a recent technological achievement in deep learning.These generative models are computationally expensive because a GAN consists of two neural networks and trains on enormous datasets.Typically,a GAN is trained on a single server.Conventional deep learning accelerator designs are challenged by the unique properties of GAN,like the enormous computation stages with non-traditional convolution layers.This work addresses the issue of distributing GANs so that they can train on datasets distributed over many TPUs(Tensor Processing Unit).Distributed learning training accelerates the learning process and decreases computation time.In this paper,the Generative Adversarial Network is accelerated using the distributed multi-core TPU in distributed data-parallel synchronous model.For adequate acceleration of the GAN network,the data parallel SGD(Stochastic Gradient Descent)model is implemented in multi-core TPU using distributed TensorFlow with mixed precision,bfloat16,and XLA(Accelerated Linear Algebra).The study was conducted on the MNIST dataset for varying batch sizes from 64 to 512 for 30 epochs in distributed SGD in TPU v3 with 128×128 systolic array.An extensive batch technique is implemented in bfloat16 to decrease the storage cost and speed up floating-point computations.The accelerated learning curve for the generator and discriminator network is obtained.The training time was reduced by 79%by varying the batch size from 64 to 512 in multi-core TPU.展开更多
This paper studies the nonlinear mixed problem for a class of symmetric hyperbolic systems with the boundary condition satisfying the dissipative condition about discontinuous data in higher dimension spaces, establis...This paper studies the nonlinear mixed problem for a class of symmetric hyperbolic systems with the boundary condition satisfying the dissipative condition about discontinuous data in higher dimension spaces, establishes the local existence theorem by using the method of a prior estimates, and obtains the structure of singularities of the solutions of such problems.展开更多
The three dimensional variable cross-section roll forming is a kind of new metal forming technol- ogy which combines large forming force, multi-axis linkage movement and space synergic movement, and the sequential syn...The three dimensional variable cross-section roll forming is a kind of new metal forming technol- ogy which combines large forming force, multi-axis linkage movement and space synergic movement, and the sequential synergic movement of the ganged roller group is used to complete the metal sheet forming according to the shape of the complicated and variable forming part data. The control system should meet the demands of quick response to the test requirements of the product part. A new kind of real time data driving multi-axis linkage and synergic movement control strategy of 3D roll forming is put forward in the paper. In the new control strategy, the forming data are automatically generated according to the shape of the parts, and the multi-axis linkage movement together with cooperative motion among the six stands of the 3D roll forming machine is driven by the real-time information, and the control nodes are also driven by the forming data. The new control strategy is applied to a 48 axis 3D roll forming machine developed by our research center, and the control servo period is less than 10ms. A forming experiment of variable cross section part is carried out, and the forming preci- sion is better than ＋ 0.5mm by the control strategy. The result of the experiment proves that the control strategy has significant potentiality for the development of 3D roll forming production line with large scale, multi-axis ganged and svner^ic movement展开更多
In this review, we highlight some recent methodological and theoretical develop- ments in estimation and testing of large panel data models with cross-sectional dependence. The paper begins with a discussion of issues...In this review, we highlight some recent methodological and theoretical develop- ments in estimation and testing of large panel data models with cross-sectional dependence. The paper begins with a discussion of issues of cross-sectional dependence, and introduces the concepts of weak and strong cross-sectional dependence. Then, the main attention is primarily paid to spatial and factor approaches for modeling cross-sectional dependence for both linear and nonlinear （nonparametric and semiparametric） panel data models. Finally, we conclude with some speculations on future research directions.展开更多
Tunnel deformation monitoring is a crucial task to evaluate tunnel stability during the metro operation period.Terrestrial Laser Scanning(TLS)can collect high density and high accuracy point cloud data in a few minute...Tunnel deformation monitoring is a crucial task to evaluate tunnel stability during the metro operation period.Terrestrial Laser Scanning(TLS)can collect high density and high accuracy point cloud data in a few minutes as an innovation technique,which provides promising applications in tunnel deformation monitoring.Here,an efficient method for extracting tunnel cross-sections and convergence analysis using dense TLS point cloud data is proposed.First,the tunnel orientation is determined using principal component analysis(PCA)in the Euclidean plane.Two control points are introduced to detect and remove the unsuitable points by using point cloud division and then the ground points are removed by defining an elevation value width of 0.5 m.Next,a z-score method is introduced to detect and remove the outlies.Because the tunnel cross-section’s standard shape is round,the circle fitting is implemented using the least-squares method.Afterward,the convergence analysis is made at the angles of 0°,30°and 150°.The proposed approach’s feasibility is tested on a TLS point cloud of a Nanjing subway tunnel acquired using a FARO X330 laser scanner.The results indicate that the proposed methodology achieves an overall accuracy of 1.34 mm,which is also in agreement with the measurements acquired by a total station instrument.The proposed methodology provides new insights and references for the applications of TLS in tunnel deformation monitoring,which can also be extended to other engineering applications.展开更多
In this article, robust generalized estimating equation for the analysis of partial linear mixed model for longitudinal data is used. The authors approximate the nonparametric function by a regression spline. Under so...In this article, robust generalized estimating equation for the analysis of partial linear mixed model for longitudinal data is used. The authors approximate the nonparametric function by a regression spline. Under some regular conditions, the asymptotic properties of the estimators are obtained. To avoid the computation of high-dimensional integral, a robust Monte Carlo Newton-Raphson algorithm is used. Some simulations are carried out to study the performance of the proposed robust estimators. In addition, the authors also study the robustness and the efficiency of the proposed estimators by simulation. Finally, two real longitudinal data sets are analyzed.展开更多
Outlier detection is a key research area in data mining technologies,as outlier detection can identify data inconsistent within a data set.Outlier detection aims to find an abnormal data size from a large data size an...Outlier detection is a key research area in data mining technologies,as outlier detection can identify data inconsistent within a data set.Outlier detection aims to find an abnormal data size from a large data size and has been applied in many fields including fraud detection,network intrusion detection,disaster prediction,medical diagnosis,public security,and image processing.While outlier detection has been widely applied in real systems,its effectiveness is challenged by higher dimensions and redundant data attributes,leading to detection errors and complicated calculations.The prevalence of mixed data is a current issue for outlier detection algorithms.An outlier detection method of mixed data based on neighborhood combinatorial entropy is studied to improve outlier detection performance by reducing data dimension using an attribute reduction algorithm.The significance of attributes is determined,and fewer influencing attributes are removed based on neighborhood combinatorial entropy.Outlier detection is conducted using the algorithm of local outlier factor.The proposed outlier detection method can be applied effectively in numerical and mixed multidimensional data using neighborhood combinatorial entropy.In the experimental part of this paper,we give a comparison on outlier detection before and after attribute reduction.In a comparative analysis,we give results of the enhanced outlier detection accuracy by removing the fewer influencing attributes in numerical and mixed multidimensional data.展开更多
Utilizing the 45 a European Centre for Medium-Range Weather Forecasts （ECMWF） reanalysis wave da- ta （ERA-40）, the long-term trend of the sea surface wind speed and （wind wave, swell, mixed wave） wave height in ...Utilizing the 45 a European Centre for Medium-Range Weather Forecasts （ECMWF） reanalysis wave da- ta （ERA-40）, the long-term trend of the sea surface wind speed and （wind wave, swell, mixed wave） wave height in the global ocean at grid point 1.5°× 1.5° during the last 44 a is analyzed. It is discovered that a ma- jority of global ocean swell wave height exhibits a significant linear increasing trend （2-8 cm/decade）, the distribution of annual linear trend of the significant wave height （SWH） has good consistency with that of the swell wave height. The sea surface wind speed shows an annually linear increasing trend mainly con- centrated in the most waters of Southern Hemisphere westerlies, high latitude of the North Pacific, Indian Ocean north of 30°S, the waters near the western equatorial Pacific and low latitudes of the Atlantic waters, and the annually linear decreasing mainly in central and eastern equator of the Pacific, Juan. Fernandez Archipelago, the waters near South Georgia Island in the Atlantic waters. The linear variational distribution characteristic of the wind wave height is similar to that of the sea surface wind speed. Another find is that the swell is dominant in the mixed wave, the swell index in the central ocean is generally greater than that in the offshore, and the swell index in the eastern ocean coast is greater than that in the western ocean inshore, and in year-round hemisphere westerlies the swell index is relatively low.展开更多
The integration of Mixed Reality(MR)technology into Autonomous Vehicles(AVs)has ushered in a new era for the automotive industry,offering heightened safety,convenience,and passenger comfort.However,the substantial and...The integration of Mixed Reality(MR)technology into Autonomous Vehicles(AVs)has ushered in a new era for the automotive industry,offering heightened safety,convenience,and passenger comfort.However,the substantial and varied data generated by MR-Connected AVs(MR-CAVs),encompassing both highly dynamic and static information,presents formidable challenges for efficient data management and retrieval.In this paper,we formulate our indexing problem as a constrained optimization problem,with the aim of maximizing the utility function that represents the overall performance of our indexing system.This optimization problem encompasses multiple decision variables and constraints,rendering it mathematically infeasible to solve directly.Therefore,we propose a heuristic algorithm to address the combinatorial complexity of the problem.Our heuristic indexing algorithm efficiently divides data into highly dynamic and static categories,distributing the index across Roadside Units(RSUs)and optimizing query processing.Our approach takes advantage of the computational capabilities of edge servers or RSUs to perform indexing operations,thereby shifting the burden away from the vehicles themselves.Our algorithm strategically places data in the cache,optimizing cache hit rate and space utilization while reducing latency.The quantitative evaluation demonstrates the superiority of our proposed scheme,with significant reductions in latency(averaging 27%-49.25%),a 30.75%improvement in throughput,a 22.50%enhancement in cache hit rate,and a 32%-50.75%improvement in space utilization compared to baseline schemes.展开更多
MODIS （Moderate Resolution Imaging Spectroradiometer） is a key instrument aboard the Terra （EOS AM） and Aqua （EOS PM） satellites. Linear spectral mixture models are applied to MOIDS data for the sub-pixel classi...MODIS （Moderate Resolution Imaging Spectroradiometer） is a key instrument aboard the Terra （EOS AM） and Aqua （EOS PM） satellites. Linear spectral mixture models are applied to MOIDS data for the sub-pixel classification of land covers. Shaoxing county of Zhejiang Province in China was chosen to be the study site and early rice was selected as the study crop. The derived proportions of land covers from MODIS pixel using linear spectral mixture models were compared with unsupervised classification derived from TM data acquired on the same day, which implies that MODIS data could be used as satellite data source for rice cultivation area estimation, possibly rice growth monitoring and yield forecasting on the regional scale.展开更多
Data envelopment analysis(DEA) model is widely used to evaluate the relative efficiency of producers. It is a kind of objective decision method with multiple indexes. However, the two basic models frequently used at p...Data envelopment analysis(DEA) model is widely used to evaluate the relative efficiency of producers. It is a kind of objective decision method with multiple indexes. However, the two basic models frequently used at present, the C2R model and the C2GS2 model have limitations when used alone,resulting in evaluations that are often unsatisfactory. In order to solve this problem, a mixed DEA model is built and is used to evaluate the validity of the business efficiency of listed companies. An explanation of how to use this mixed DEA model is offered and its feasibility is verified.展开更多
The expected mean squares for unbalanced mixed effect interactive model were derived using Brute Force Method. From the expected mean squares, there are no obvious denominators for testing for the main effects when th...The expected mean squares for unbalanced mixed effect interactive model were derived using Brute Force Method. From the expected mean squares, there are no obvious denominators for testing for the main effects when the factors are mixed. An expression for F-test for testing for the main effects was derived which was proved to be unbiased.展开更多
The scientists are dedicated to studying the detection of Alzheimer’s disease onset to find a cure, or at the very least, medication that can slow the progression of the disease. This article explores the effectivene...The scientists are dedicated to studying the detection of Alzheimer’s disease onset to find a cure, or at the very least, medication that can slow the progression of the disease. This article explores the effectiveness of longitudinal data analysis, artificial intelligence, and machine learning approaches based on magnetic resonance imaging and positron emission tomography neuroimaging modalities for progression estimation and the detection of Alzheimer’s disease onset. The significance of feature extraction in highly complex neuroimaging data, identification of vulnerable brain regions, and the determination of the threshold values for plaques, tangles, and neurodegeneration of these regions will extensively be evaluated. Developing automated methods to improve the aforementioned research areas would enable specialists to determine the progression of the disease and find the link between the biomarkers and more accurate detection of Alzheimer’s disease onset.展开更多
文摘Deep neural networks are gaining importance and popularity in applications and services.Due to the enormous number of learnable parameters and datasets,the training of neural networks is computationally costly.Parallel and distributed computation-based strategies are used to accelerate this training process.Generative Adversarial Networks(GAN)are a recent technological achievement in deep learning.These generative models are computationally expensive because a GAN consists of two neural networks and trains on enormous datasets.Typically,a GAN is trained on a single server.Conventional deep learning accelerator designs are challenged by the unique properties of GAN,like the enormous computation stages with non-traditional convolution layers.This work addresses the issue of distributing GANs so that they can train on datasets distributed over many TPUs(Tensor Processing Unit).Distributed learning training accelerates the learning process and decreases computation time.In this paper,the Generative Adversarial Network is accelerated using the distributed multi-core TPU in distributed data-parallel synchronous model.For adequate acceleration of the GAN network,the data parallel SGD(Stochastic Gradient Descent)model is implemented in multi-core TPU using distributed TensorFlow with mixed precision,bfloat16,and XLA(Accelerated Linear Algebra).The study was conducted on the MNIST dataset for varying batch sizes from 64 to 512 for 30 epochs in distributed SGD in TPU v3 with 128×128 systolic array.An extensive batch technique is implemented in bfloat16 to decrease the storage cost and speed up floating-point computations.The accelerated learning curve for the generator and discriminator network is obtained.The training time was reduced by 79%by varying the batch size from 64 to 512 in multi-core TPU.
文摘This paper studies the nonlinear mixed problem for a class of symmetric hyperbolic systems with the boundary condition satisfying the dissipative condition about discontinuous data in higher dimension spaces, establishes the local existence theorem by using the method of a prior estimates, and obtains the structure of singularities of the solutions of such problems.
基金Supported by National Key Technology R&D Program(No.2011BAG03B03)
文摘The three dimensional variable cross-section roll forming is a kind of new metal forming technol- ogy which combines large forming force, multi-axis linkage movement and space synergic movement, and the sequential synergic movement of the ganged roller group is used to complete the metal sheet forming according to the shape of the complicated and variable forming part data. The control system should meet the demands of quick response to the test requirements of the product part. A new kind of real time data driving multi-axis linkage and synergic movement control strategy of 3D roll forming is put forward in the paper. In the new control strategy, the forming data are automatically generated according to the shape of the parts, and the multi-axis linkage movement together with cooperative motion among the six stands of the 3D roll forming machine is driven by the real-time information, and the control nodes are also driven by the forming data. The new control strategy is applied to a 48 axis 3D roll forming machine developed by our research center, and the control servo period is less than 10ms. A forming experiment of variable cross section part is carried out, and the forming preci- sion is better than ＋ 0.5mm by the control strategy. The result of the experiment proves that the control strategy has significant potentiality for the development of 3D roll forming production line with large scale, multi-axis ganged and svner^ic movement
基金Supported by the National Natural Science Foundation of China(71131008(Key Project)and 71271179)
文摘In this review, we highlight some recent methodological and theoretical develop- ments in estimation and testing of large panel data models with cross-sectional dependence. The paper begins with a discussion of issues of cross-sectional dependence, and introduces the concepts of weak and strong cross-sectional dependence. Then, the main attention is primarily paid to spatial and factor approaches for modeling cross-sectional dependence for both linear and nonlinear （nonparametric and semiparametric） panel data models. Finally, we conclude with some speculations on future research directions.
基金National Natural Science Foundation of China(No.41801379)Fundamental Research Funds for the Central Universities(No.2019B08414)National Key R&D Program of China(No.2016YFC0401801)。
文摘Tunnel deformation monitoring is a crucial task to evaluate tunnel stability during the metro operation period.Terrestrial Laser Scanning(TLS)can collect high density and high accuracy point cloud data in a few minutes as an innovation technique,which provides promising applications in tunnel deformation monitoring.Here,an efficient method for extracting tunnel cross-sections and convergence analysis using dense TLS point cloud data is proposed.First,the tunnel orientation is determined using principal component analysis(PCA)in the Euclidean plane.Two control points are introduced to detect and remove the unsuitable points by using point cloud division and then the ground points are removed by defining an elevation value width of 0.5 m.Next,a z-score method is introduced to detect and remove the outlies.Because the tunnel cross-section’s standard shape is round,the circle fitting is implemented using the least-squares method.Afterward,the convergence analysis is made at the angles of 0°,30°and 150°.The proposed approach’s feasibility is tested on a TLS point cloud of a Nanjing subway tunnel acquired using a FARO X330 laser scanner.The results indicate that the proposed methodology achieves an overall accuracy of 1.34 mm,which is also in agreement with the measurements acquired by a total station instrument.The proposed methodology provides new insights and references for the applications of TLS in tunnel deformation monitoring,which can also be extended to other engineering applications.
基金the Natural Science Foundation of China(10371042,10671038)
文摘In this article, robust generalized estimating equation for the analysis of partial linear mixed model for longitudinal data is used. The authors approximate the nonparametric function by a regression spline. Under some regular conditions, the asymptotic properties of the estimators are obtained. To avoid the computation of high-dimensional integral, a robust Monte Carlo Newton-Raphson algorithm is used. Some simulations are carried out to study the performance of the proposed robust estimators. In addition, the authors also study the robustness and the efficiency of the proposed estimators by simulation. Finally, two real longitudinal data sets are analyzed.
基金The authors would like to acknowledge the support of Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai)(SML2020SP007)The paper is supported under the National Natural Science Foundation of China(Nos.61772280 and 62072249).
文摘Outlier detection is a key research area in data mining technologies,as outlier detection can identify data inconsistent within a data set.Outlier detection aims to find an abnormal data size from a large data size and has been applied in many fields including fraud detection,network intrusion detection,disaster prediction,medical diagnosis,public security,and image processing.While outlier detection has been widely applied in real systems,its effectiveness is challenged by higher dimensions and redundant data attributes,leading to detection errors and complicated calculations.The prevalence of mixed data is a current issue for outlier detection algorithms.An outlier detection method of mixed data based on neighborhood combinatorial entropy is studied to improve outlier detection performance by reducing data dimension using an attribute reduction algorithm.The significance of attributes is determined,and fewer influencing attributes are removed based on neighborhood combinatorial entropy.Outlier detection is conducted using the algorithm of local outlier factor.The proposed outlier detection method can be applied effectively in numerical and mixed multidimensional data using neighborhood combinatorial entropy.In the experimental part of this paper,we give a comparison on outlier detection before and after attribute reduction.In a comparative analysis,we give results of the enhanced outlier detection accuracy by removing the fewer influencing attributes in numerical and mixed multidimensional data.
基金The National Basic Research Program of China under contract No.2012CB957803
文摘Utilizing the 45 a European Centre for Medium-Range Weather Forecasts （ECMWF） reanalysis wave da- ta （ERA-40）, the long-term trend of the sea surface wind speed and （wind wave, swell, mixed wave） wave height in the global ocean at grid point 1.5°× 1.5° during the last 44 a is analyzed. It is discovered that a ma- jority of global ocean swell wave height exhibits a significant linear increasing trend （2-8 cm/decade）, the distribution of annual linear trend of the significant wave height （SWH） has good consistency with that of the swell wave height. The sea surface wind speed shows an annually linear increasing trend mainly con- centrated in the most waters of Southern Hemisphere westerlies, high latitude of the North Pacific, Indian Ocean north of 30°S, the waters near the western equatorial Pacific and low latitudes of the Atlantic waters, and the annually linear decreasing mainly in central and eastern equator of the Pacific, Juan. Fernandez Archipelago, the waters near South Georgia Island in the Atlantic waters. The linear variational distribution characteristic of the wind wave height is similar to that of the sea surface wind speed. Another find is that the swell is dominant in the mixed wave, the swell index in the central ocean is generally greater than that in the offshore, and the swell index in the eastern ocean coast is greater than that in the western ocean inshore, and in year-round hemisphere westerlies the swell index is relatively low.
文摘The integration of Mixed Reality(MR)technology into Autonomous Vehicles(AVs)has ushered in a new era for the automotive industry,offering heightened safety,convenience,and passenger comfort.However,the substantial and varied data generated by MR-Connected AVs(MR-CAVs),encompassing both highly dynamic and static information,presents formidable challenges for efficient data management and retrieval.In this paper,we formulate our indexing problem as a constrained optimization problem,with the aim of maximizing the utility function that represents the overall performance of our indexing system.This optimization problem encompasses multiple decision variables and constraints,rendering it mathematically infeasible to solve directly.Therefore,we propose a heuristic algorithm to address the combinatorial complexity of the problem.Our heuristic indexing algorithm efficiently divides data into highly dynamic and static categories,distributing the index across Roadside Units(RSUs)and optimizing query processing.Our approach takes advantage of the computational capabilities of edge servers or RSUs to perform indexing operations,thereby shifting the burden away from the vehicles themselves.Our algorithm strategically places data in the cache,optimizing cache hit rate and space utilization while reducing latency.The quantitative evaluation demonstrates the superiority of our proposed scheme,with significant reductions in latency(averaging 27%-49.25%),a 30.75%improvement in throughput,a 22.50%enhancement in cache hit rate,and a 32%-50.75%improvement in space utilization compared to baseline schemes.
文摘MODIS （Moderate Resolution Imaging Spectroradiometer） is a key instrument aboard the Terra （EOS AM） and Aqua （EOS PM） satellites. Linear spectral mixture models are applied to MOIDS data for the sub-pixel classification of land covers. Shaoxing county of Zhejiang Province in China was chosen to be the study site and early rice was selected as the study crop. The derived proportions of land covers from MODIS pixel using linear spectral mixture models were compared with unsupervised classification derived from TM data acquired on the same day, which implies that MODIS data could be used as satellite data source for rice cultivation area estimation, possibly rice growth monitoring and yield forecasting on the regional scale.
基金Supported by Commission of Science Technology and Industry for National Defense(No, C192005C001)
文摘Data envelopment analysis(DEA) model is widely used to evaluate the relative efficiency of producers. It is a kind of objective decision method with multiple indexes. However, the two basic models frequently used at present, the C2R model and the C2GS2 model have limitations when used alone,resulting in evaluations that are often unsatisfactory. In order to solve this problem, a mixed DEA model is built and is used to evaluate the validity of the business efficiency of listed companies. An explanation of how to use this mixed DEA model is offered and its feasibility is verified.
文摘The expected mean squares for unbalanced mixed effect interactive model were derived using Brute Force Method. From the expected mean squares, there are no obvious denominators for testing for the main effects when the factors are mixed. An expression for F-test for testing for the main effects was derived which was proved to be unbiased.
文摘The scientists are dedicated to studying the detection of Alzheimer’s disease onset to find a cure, or at the very least, medication that can slow the progression of the disease. This article explores the effectiveness of longitudinal data analysis, artificial intelligence, and machine learning approaches based on magnetic resonance imaging and positron emission tomography neuroimaging modalities for progression estimation and the detection of Alzheimer’s disease onset. The significance of feature extraction in highly complex neuroimaging data, identification of vulnerable brain regions, and the determination of the threshold values for plaques, tangles, and neurodegeneration of these regions will extensively be evaluated. Developing automated methods to improve the aforementioned research areas would enable specialists to determine the progression of the disease and find the link between the biomarkers and more accurate detection of Alzheimer’s disease onset.
