检索结果-内蒙古大学图书馆

Credit assignment for trained neural networks based on Koopman operator theory

Frontiers of Computer Science 2024年第1期18卷 251-253页

作者： Zhen LIANG Changyuan ZHAO Wanwei LIU Bai XUE Wenjing YANG Zhengbin PANG Institute for Quantum Information&State Key Laboratory of High Performance Computing National University of Defense TechnologyChangsha 410000China Institute of Software Chinese Academy of SciencesBeijing 100190China College of Computer Science and Technology National University of Defense TechnologyChangsha 410000China

1 Introduction Artificial neural networks(ANNs,also NNs)have recently emerged as leading candidate models for deep learning,popularly used in various areas[1–3].Behind the enormous success,ANNs are generally with complicated structures,there being an intricate data flow through multiple linear or nonlinear components between the input layer and the output ***,it is pressing to evaluate how much a specific component contributes to the final output,termed the Credit Assignment Problem(CAP)[4]in this paper.

关键词： layer assignment operator

来源：评论

学校读者我要写书评

暂无评论

Evaluating performance Portability of SYCL and Kokkos: A Case Study on LBM Simulations 21

Evaluating Performance Portability of SYCL and Kokkos: A Cas...

引用

21st IEEE International Symposium on Parallel and Distributed Processing with Applications, 13th IEEE International Conference on Big Data and Cloud computing, 16th IEEE International Conference on Social computing and Networking and 13th International Conference on Sustainable computing and Communications, ISPA/BDCloud/SocialCom/SustainCom 2023

作者： Ding, Yue Xu, Chuanfu Qiu, Haozhong Wang, Qingsong Dai, Weixi Lin, Yongzhen Che, Yonggang National University of Defense Technology Institute for Quantum Information State Key Laboratory of High Performance Computing Changsha China

ISBN: (纸本)9798350329223

Since modern high performance computing systems are evolving towards diverse and heterogeneous architectures, the emergence of high-level portable programming models leads to a particular focus on performance portability. In this paper, we evaluate the performance portability and explore performance optimization methods for two portable programming models SYCL and Kokkos. We take an open-source multi-phase Lattice Boltzmann Method (LBM) flow simulation code as a case study and implement portable versions with different optimizations. Then we compare our portable implementations with engineer-tuned OpenMP and CUDA versions on Intel CPUs and NVIDIA GPUs. Experimental results show that both SYCL and Kokkos can deliver superior performance than traditional programming models, but the best performance of the portable versions depends heavily on platform-specific optimizations. There is no single implementation that can achieve the best performance on both CPUs and GPUs. Consequently, we conclude that the performance portability still needs to be further improved for both SYCL and Kokkos. In addition, we present a comparative analysis of different optimization methods that qualify the performance enhancement when using SYCL and Kokkos on CPUs and GPUs. Our work offers valuable references for the development of both portable programming models and applications. © 2023 IEEE.

关键词： Program processors

来源：评论

学校读者我要写书评

暂无评论

Autism Spectrum Disorder Classification with Interpretability in Children Based on Structural MRI Features Extracted Using Contrastive Variational Autoencoder

引用

Big Data Mining and Analytics 2024年第3期7卷 781-793页

作者： Ruimin Ma Ruitao Xie Yanlin Wang Jintao Meng Yanjie Wei Yunpeng Cai Wenhui Xi Yi Pan Shenzhen Institute of Advanced Technology Chinese Academy of SciencesShenzhen 518055China University of Chinese Academy of Sciences Beijing 100049China Shenzhen Key Laboratory of Intelligent Bioinformatics and with Center for High Performance ComputingShenzhen Institute of Advanced TechnologyChinese Academy of SciencesShenzhen 518055China

Autism Spectrum Disorder(ASD)is a highly disabling mental disease that brings significant impairments of social interaction ability to the patients,making early screening and intervention of ASD *** the development of the machine learning and neuroimaging technology,extensive research has been conducted on machine classification of ASD based on structural Magnetic Resonance Imaging(s-MRI).However,most studies involve with datasets where participants'age are above 5 and lack *** this paper,we propose a machine learning method for ASD classification in children with age range from 0.92 to 4.83 years,based on s-MRI features extracted using Contrastive Variational AutoEncoder(CVAE).78 s-MRIs,collected from Shenzhen Children's Hospital,are used for training CVAE,which consists of both ASD-specific feature channel and common-shared feature *** ASD participants represented by ASD-specific features can be easily discriminated from Typical Control(TC)participants represented by the common-shared *** case of degraded predictive accuracy when data size is extremely small,a transfer learning strategy is proposed here as a potential ***,we conduct neuroanatomical interpretation based on the correlation between s-MRI features extracted from CVAE and surface area of different cortical regions,which discloses potential biomarkers that could help target treatments of ASD in the future.

关键词： Autism Spectrum Disorder(ASD)classification Contrastive Variational AutoEncoder(CVAE) transfer learning neuroanatomical interpretation

来源：评论

学校读者我要写书评

暂无评论

Thermal conductivity of GeTe crystals based on machine learning potentials

引用

Chinese Physics B 2024年第4期33卷 104-107页

作者：张健张昊春李伟峰张刚 School of Energy Science and Engineering Harbin Institute of TechnologyHarbin 150001China Institute of High Performance Computing Agency for ScienceTechnology and ResearchSingapore 138632Singapore School of Physics&State Key Laboratory of Crystal Materials Shandong UniversityJinan 250100China

GeTe has attracted extensive research interest for thermoelectric *** this paper,we first train a neuroevolution potential(NEP)based on a dataset constructed by ab initio molecular dynamics,with the Gaussian approximation potential(GAP)as a *** phonon density of states is then calculated by two machine learning potentials and compared with density functional theory results,with the GAP potential having higher ***,the thermal conductivity of a GeTe crystal at 300 K is calculated by the equilibrium molecular dynamics method using both machine learning potentials,and both of them are in good agreement with the experimental results;however,the calculation speed when using the NEP potential is about 500 times faster than when using the GAP ***,the lattice thermal conductivity in the range of 300 K-600 K is calculated using the NEP *** lattice thermal conductivity decreases as the temperature increases due to the phonon anharmonic *** study provides a theoretical tool for the study of the thermal conductivity of GeTe.

关键词： machine learning potentials thermal conductivity molecular dynamics

来源：评论

学校读者我要写书评

暂无评论

New Finite Difference Mapped WENO Schemes with Increasingly high Order of Accuracy

引用

Communications on Applied Mathematics and Computation 2023年第1期5卷 64-96页

作者： Jun Zhu Jianxian Qiu College of Science Nanjing University of Aeronautics and AstronauticsNanjing 210016JiangsuChina School of Mathematical Sciences and Fujian Provincial Key Laboratory of Mathematical Modeling and High-Performance Scientific Computing Xiamen UniversityXiamen 361005FujianChina

In this paper,a new type of finite difference mapped weighted essentially non-oscillatory(MWENO)schemes with unequal-sized stencils,such as the seventh-order and ninthorder versions,is constructed for solving hyperbolic conservation *** the purpose of designing increasingly high-order finite difference WENO schemes,the equal-sized stencils are becoming more and more *** more we use wider candidate stencils,the bigger the probability of discontinuities lies in all ***,one innovation of these new WENO schemes is to introduce a new splitting stencil methodology to divide some fourpoint or five-point stencils into several smaller three-point *** the usage of this new methodology in high-order spatial reconstruction procedure,we get different degree polynomials defined on these unequal-sized stencils,and calculate the linear weights,smoothness indicators,and nonlinear weights as specified in Jiang and Shu(***.126:202228,1996).Since the difference between the nonlinear weights and the linear weights is too big to keep the optimal order of accuracy in smooth regions,another crucial innovation is to present the new mapping functions which are used to obtain the mapped nonlinear weights and decrease the difference quantity between the mapped nonlinear weights and the linear weights,so as to keep the optimal order of accuracy in smooth *** new MWENO schemes can also be applied to compute some extreme examples,such as the double rarefaction wave problem,the Sedov blast wave problem,and the Leblanc problem with a normal CFL *** numerical results are provided to illustrate the good performance of the new finite difference MWENO schemes.

关键词： Finite difference Mapped WENO scheme Mapping function Mapped nonlinear weight Unequal-sized stencil Extreme example

来源：评论

学校读者我要写书评

暂无评论

Hot Deformation Behavior and Microstructures Evolution of GNP-Reinforced Fine-Grained Mg Composites

引用

Acta Metallurgica Sinica(English Letters) 2024年第3期37卷 407-424页

作者： Hengrui Hu Jiayu Qin Yunpeng Zhu Jinhui Wang Xiaoqiang Li Peipeng Jin Qinghai Provincial Key Laboratory of New Light Alloys Qinghai UniversityXining 810016China Qinghai Provincial Engineering Research Center of High Performance Light Metal Alloys and Forming Qinghai UniversityXining 810016China Department of Mechanical Engineering Tsinghua UniversityBeijing 100084China School of Material Science and Engineering Lanzhou University of TechnologyLanzhou 730050China

Graphene nanoplates(GNPs)-reinforced magnesium matrix composites have been attracted great ***,knowledge is lack for the hot deformation behavior of GNP-reinforced magnesium(GNPs/Mg)*** this study,the fine-grained GNPs/Mg composite was fabricated by powder metallurgy process followed by *** hot deformation behavior,microstructure evolution and dynamic recrystallization(DRX)mechanism of fine-grained GNPs/Mg composite were investigated by hot compression test and electron back-scatter diffraction(EBSD).The hot compression tests of the composite were conducted at temperatures between 423 and 573 K with the strain rates from 0.001 to 1 s^(-1).The strain compensated power law equation was established to describe the hot deformation behavior of the *** stress exponent and activation energy of the composite are 7.76 and 83.23 kJ/mol,respectively,suggesting that the deformation mechanism is grain boundary slip controlled dislocation climb *** abnormally high stress exponent and activation energy are unattainable in the composite due to the fine grain size of the composites and the absence of Zener pinning and Orowan effects of GNPs *** grain size increases with the decrease in Zener-Hollomn(Z)parameter,which can be well fitted by power-law *** the increase in grain size and decrease in Z parameter,the geometrically necessary dislocation density decreases,which shows the approximately power-law relationship.A random and weak texture was formed after hot *** discontinuous dynamic recrystallization and continuous dynamic recrystallization mechanism dominated the DRX behavior at 473 K/0.001 s^(-1) and 573 K/0.001 s^(-1),respectively.

关键词： GNPs Mg composite Hot deformation behavior Constitutive equations Microstructure evolution Dynamic recrystallization

来源：评论

学校读者我要写书评

暂无评论

Asymptotic stability of explicit infinite energy blowup solutions of the 3D incompressible Navier-Stokes equations

引用

Science China Mathematics 2023年第11期66卷 2523-2544页

作者： Fangyu Han Zhong Tan School of Mathematical Sciences Xiamen UniversityXiamen 361005China Fujian Provincial Key Laboratory on Mathematical Modeling and High-Performance Scientific Computing Xiamen UniversityXiamen UniversityXiamen 361005China Shenzhen Research Institute of Xiamen University Shenzhen 518057China

In this paper,we study the dynamical stability of a family of explicit blowup solutions of the threedimensional(3D)incompressible Navier-Stokes(NS)equations with smooth initial values,which is constructed in Guo et al.(2008).This family of solutions has finite energy in any bounded domain of R3,but unbounded energy in *** on similarity coordinates,energy estimates and the Nash-Moser-H?rmander iteration scheme,we show that these solutions are asymptotically stable in the backward light-cone of the ***,the result shows the existence of local energy blowup solutions to the 3D incompressible NS equations with growing ***,the result also shows that in the absence of physical boundaries,the viscous vanishing limit of the solutions does not satisfy the 3D incompressible Euler equations.

关键词： Navier-Stokes equations asymptotic stability blowup solution infinite energy Nash-Moser-Hormander iteration scheme zero-viscosity limit

来源：评论

学校读者我要写书评

暂无评论

SPM-GCN: An adaptive reordering algorithm for sparse LU factorization via GCN 29

SPM-GCN: An adaptive reordering algorithm for sparse LU fact...

引用

29th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2023

作者： Tian, Min Liu, Huazeng Wang, Junjie Liu, Qi Du, Wei Zhang, Zanjun China Shandong Fundamental Research Center for Computer Science Jinan Key Laboratory of High Performance Computing Shandong Provincial Key Laboratory of Computer Networks China Xidian University Shaanxi Key Laboratory of Large Scale Electromagnetic Computing Xi'an710071 China

ISBN: (纸本)9798350330717

Sparse LU factorization is a critical kernel in scientific computing and engineering applications. A better nonzero pattern of sparse matrixes can accelerate LU factorization by reordering. Traditionally it's difficult to predict which non-zero pattern is optimal for a sparse matrix. In this paper, we proposed a graph convolutional neural network (GCN) for adaptively selecting the optimal reordering algorithm from five candidate reordering algorithms in the Matrix preprocessing step, referred to as SPM-GCN. After using our SPM-GCN, the average numerical factorization time outperformed the other five algorithms, and the average numerical factorization time was reduced by 17.3% compared with the default method of swSuperLU. © 2023 IEEE.

关键词： Graph Convolutional Network reordering algorithm sparse linear system solver Sparse LU factorization

来源：评论

学校读者我要写书评

暂无评论

Qualitative and Quantitative Model Checking Against Recurrent Neural Networks

引用

Journal of Computer Science & technology 2024年第6期39卷 1292-1311页

作者： Zhen Liang Wan-Wei Liu Fu Song Bai Xue Wen-Jing Yang Ji Wang Zheng-Bin Pang Institute for Quantum Information&State Key Laboratory of High Performance Computing College of Computer Science and TechnologyNational University of Defense TechnologyChangsha 410073China College of Computer Science and Technology National University of Defense TechnologyChangsha 410073China School of Information Science and Technology ShanghaiTech UniversityShanghai 201210China Institute of Software Chinese Academy of SciencesBeijing 100190China

Recurrent neural networks (RNNs) have been heavily used in applications relying on sequence data such as time series and natural languages. As a matter of fact, their behaviors lack rigorous quality assurance due to the black-box nature of deep learning. It is an urgent and challenging task to formally reason about the behaviors of RNNs. To this end, we first present an extension of linear-time temporal logic to reason about properties with respect to RNNs, such as local robustness, reachability, and some temporal properties. Based on the proposed logic, we formalize the verification obligation as a Hoare-like triple, from both qualitative and quantitative perspectives. The former concerns whether all the outputs resulting from the inputs fulfilling the pre-condition satisfy the post-condition, whereas the latter is to compute the probability that the post-condition is satisfied on the premise that the inputs fulfill the pre-condition. To tackle these problems, we develop a systematic verification framework, mainly based on polyhedron propagation, dimension-preserving abstraction, and the Monte Carlo sampling. We also implement our algorithm with a prototype tool and conduct experiments to demonstrate its feasibility and efficiency.

关键词： recurrent neural network model checking temporal logic qualitative/quantitative verification

来源：评论

学校读者我要写书评

暂无评论

Ultra-low power MoS2 optoelectronic synapse with wavelength sensitivity for color target recognition

引用

Science China(Information Sciences) 2025年第4期68卷 187-196页

作者： Bo WEI Yabo CHEN Xiaotong HAN Yan KANG Bujia LIANG Cheng LI Xiaokuo YANG Liang FANG Yuanxi PENG Institute for Quantum Information & State Key Laboratory of High Performance Computing College of ComputerNational University of Defense Technology Fundamentals Department Air Force Engineering University College of Advanced Interdisciplinary Studies National University of Defense Technology Institute of Quantum Information Science and Technology College of ScienceNational University of Defense Technology College of Computer National University of Defense Technology

Optoelectronic synapses that integrate visual perception and pre-processing hold significant potential for neuromorphic vision systems（NVSs）. However, due to a lack of wavelength sensitivity, existing NVS mainly focuses on gray-scale image processing, making it challenging to recognize color images. Additionally, the high power consumption of optoelectronic synapses, compared to the 10 fJ energy consumption of biological synapses, limits their broader application. To address these challenges, an energy-efficient NVS capable of color target recognition in a noisy environment was developed,utilizing a MoS2optoelectronic synapse with wavelength sensitivity. Benefiting from the distinct photon capture capabilities of 450, 535, and 650 nm light, the optoelectronic synapse exhibits wavelength-dependent synaptic plasticity, including excitatory postsynaptic current（EPSC）, paired-pulse facilitation（PPF）, and long-term plasticity（LTP）. These properties can effectively mimic the visual memory and color discrimination functions of the human vision system. Results demonstrate that the NVS, based on MoS2optoelectronic synapses, can eliminate the color noise at the sensor level, increasing color image recognition accuracy from 50% to 90%. Importantly, the optoelectronic synapse operates at a low voltage spike of0.0005 V, consuming only 0.075 fJ per spike, surpassing the energy efficiency of both existing optoelectronic and biological synapses. This ultra-low power, color-sensitive device eliminates the need for color filters and offers great promise for future deployment in filter-free NVS.

关键词： optoelectronic synapse neuromorphic vision system color recognition 2D materials image pre-processing

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：