检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Cao, Lianghao Chen, Joshua Brennan, Michael O’Leary-Roseberry, Thomas Marzouk, Youssef Ghattas, Omar Center for Computational Science and Engineering Massachusetts Institute of Technology CambridgeMA02139 United States Department of Computing and Mathematical Sciences California Institute of Technology PasadenaCA91125 United States Oden Institute for Computational Engineering and Sciences The University of Texas at Austin 201 E. 24th Street C0200 AustinTX78712 United States Walker Department of Mechanical Engineering The University of Texas at Austin 204 E. Dean Keaton Street AustinTX78712 United States

We present LazyDINO, a transport map variational inference method for fast, scalable, and efficiently amortized solutions of high-dimensional nonlinear Bayesian inverse problems with expensive parameter-to-observable (PtO) maps. Our method consists of an offline phase in which we construct a derivative-informed neural surrogate of the PtO map using joint samples of the PtO map and its Jacobian. During the online phase, when given observational data, we seek rapid posterior approximation using surrogate-driven training of a lazy map [Brennan et al., NeurIPS, (2020)], i.e., a structure-exploiting transport map with low-dimensional nonlinearity. The trained lazy map then produces approximate posterior samples or density evaluations. Our surrogate construction is optimized for amortized Bayesian inversion using lazy map variational inference. We show that (i) the derivative-based reduced basis architecture [O’Leary-Roseberry et al., Comput. Methods Appl. Mech. Eng., 388 (2022)] minimizes the upper bound on the expected error in surrogate posterior approximation, and (ii) the derivative-informed training formulation [O’Leary-Roseberry et al., J. Comput. Phys., 496 (2024)] minimizes the expected error due to surrogate-driven transport map optimization. Our numerical results demonstrate that LazyDINO is highly efficient in cost amortization for Bayesian inversion. We observe one to two orders of magnitude reduction of offline cost for accurate posterior approximation, compared to simulation-based amortized inference via conditional transport and conventional surrogate-driven transport. In particular, LazyDINO outperforms Laplace approximation consistently using fewer than 1000 offline samples, while other amortized inference methods struggle and sometimes fail at 16,000 offline samples. © 2024, CC BY-NC-SA.

关键词： Inverse problems

来源：评论

学校读者我要写书评

暂无评论

An Improved Cuttlefish Optimization Algorithm Based Faster R-CNN for Brain Tumour Detection

An Improved Cuttlefish Optimization Algorithm Based Faster R...

引用

Smart Electronics and Communication (ICOSEC), International Conference on

作者： AR Arunarani Vedant Bhenia T Abirami N Aparna Md. Abul Ala Walid Sharada K A Department of Computational and Intelligence School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Campus Chennai Tamilnadu India Department of Computer Science and Engineering IPS Academy Institute of Engineering and Science Indore Madhya Pradesh India Information Technology Kongu Engineering College Erode Tamilnadu India Department of Biomedical Engineering Dhaanish Ahmed Institute of Technology Coimbatore Tamilnadu India Department of Computer Science and Engineering Khulna University of Engineering & Technology (KUET) Khulna Bangladesh Department of CSE HKBK College of Engineering VTU Bengaluru Karnataka India

Brain tumor is caused due to the anomalous development of tissues in the brain. Precise identification of the tumor in premature stage is essential to provide effective medication and curtail further impact of the disease in affected individuals. This present research includes four main stages for the diagnosis of brain tumor including segmentation and classification based on deep learning approaches. The preprocessing stage makes use of filters to reduce the noise impact in the brain imagery. The segmentation stage is responsible to segment the tumor portions from the images using Enhanced Region Growth algorithm. Feature selection stage utilizes Improved Cuttlefish Optimization (ICO) algorithm to extricate the optimal features from the segmented images. The final stage is classification in which Faster R-CNN (FRCNN) algorithm is employed to categorize the brain images as tumor affected or not. BRA TS-2020 dataset is used for the experimental investigation of the proposed FRCNN-ICO method. Performance of this method is assessed using parameters such as Accurate Prediction Rate (APR), Imprecise Prediction Rate (IPR) and Imprecise Denial Rate (IDR). The proposed FRCNN-ICO produces the highest APR value of 99.2% compared to the existing researches on brain tumor detection and proves to perform superior than the methods considered for comparison purposes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Mathematical and numerical study of an inverse source problem for the biharmonic wave equation

arXiv

引用

arXiv 2023年

作者： Chang, Yan Guo, Yukun Yin, Tao Zhao, Yue School of Mathematics Harbin Institute of Technology Harbin China LSEC Institute of Computational Mathematics and Scientific/Engineering Computing Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China School of Mathematics and Statistics Central China Normal University Wuhan China

In this paper, we study the inverse source problem for the biharmonic wave equation. Mathematically, we characterize the radiating sources and non-radiating sources at a fixed wavenumber. We show that a general source can be decomposed into a radiating source and a non-radiating source. The radiating source can be uniquely determined by Dirichlet boundary measurements at a fixed wavenumber. Moreover, we derive a Lipschitz stability estimate for determining the radiating source. On the other hand, the non-radiating source does not produce any scattered fields outside the support of the source function. Numerically, we propose a novel source reconstruction method based on Fourier series expansion by multi-wavenumber boundary measurements. Numerical experiments are presented to verify the accuracy and efficiency of the proposed method. Copyright © 2023, The Authors. All rights reserved.

关键词： Wave equations

来源：评论

学校读者我要写书评

暂无评论

A class of refined implicit-explicit Runge-Kutta methods with robust time adaptability and unconditional convergence for the Cahn-Hilliard model

arXiv

引用

arXiv 2024年

作者： Liao, Hong-Lin Tang, Tao Wang, Xuping Zhou, Tao School of Mathematics Nanjing University of Aeronautics and Astronautics Nanjing211106 China MIIT Nanjing211106 China School of Electrical and Computer Engineering Guangzhou Nanfang College Institute for Advanced Study BNU-HKBU United International College China Institute of Computational Mathematics and Scientific/Engineering Computing Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China

One of main obstacles in verifying the energy dissipation laws of implicit-explicit Runge-Kutta (IERK) methods for phase field equations is to establish the uniform boundedness of stage solutions without the global Lipschitz continuity assumption of nonlinear bulk. With the help of discrete orthogonal convolution kernels, an updated time-space splitting technique is developed to establish the uniform boundedness of stage solutions for a refined class of IERK methods in which the associated differentiation matrices and the average dissipation rates are always independent of the time-space discretization meshes. This makes the refined IERK methods highly advantageous in self-adaptive time-stepping procedures as some larger adaptive step-sizes in actual simulations become possible. From the perspective of optimizing the average dissipation rate, we construct some parameterized refined IERK methods up to third-order accuracy, in which the involved diagonally implicit Runge-Kutta methods for the implicit part have an explicit first stage and allow a stage-order of two such that they are not necessarily algebraically stable. Then we are able to establish, for the first time, the original energy dissipation law and the unconditional L2 norm convergence. Extensive numerical tests are presented to support our *** Codes 35K58, 65L20, 65M06, 65M12 © 2024, CC BY-NC-ND.

关键词： Runge Kutta methods

来源：评论

学校读者我要写书评

暂无评论

Revolutionizing Aquaculture Practices: IoT and Deep Learning-based Remote Monitoring for Sustainable Fish Farming

Revolutionizing Aquaculture Practices: IoT and Deep Learning...

引用

Smart Electronics and Communication (ICOSEC), International Conference on

作者： S. Prince Chelladurai G.Susan Shiny S.V Jeevitha P Swapna R. Pavithra S Arulraj Department of Computational Intelligence Faculty of Engineering and Technology School of Computing SRM Institute of Science and Technology Kattankulathur Chennai Department of information technology Sethu institute of Technology Pulloor Kariapatti Virudhunagar Tamil Nadu Department of information technology Bannari Amman Institute of Technology Erode India Department of CSE Velammal college of engineering and technology Madurai Department of ECE Rathinam Technical Campus Coimbatore India

ISBN: (数字)9798331504403

ISBN: (纸本)9798331504410

Aquaculture plays a vital role in meeting the growing demand for seafood while addressing sustainability concerns in fisheries. To optimize fish growth and ensure efficient resource utilization an innovative IoT and Deep Learning-based remote monitoring system is proposed tailored for aquaculture operations. Our system integrates sensor technology, IOT connectivity, and Deep Learning algorithms to provide real-time monitoring and control of key environmental parameters such as water temperature, pH levels, oxygen concentration, and feeding behavior. By leveraging a network of sensors deployed in aquaculture facilities, our system collects and analyzes environmental data to optimize aquaculture practices and ensure optimal conditions for fish growth. Through a user-friendly interface accessible via web or mobile applications, aqua-culturists can remotely monitor the status of their facilities and adjust operational parameters as needed. Additionally, our system utilizes Deep Learning algorithms to analyze historical data and predict optimal feeding schedules, thereby enhancing feed efficiency and fish growth rates. Field trials demonstrate the effectiveness of our IOT and Deep Learning-based monitoring system in improving fish health and productivity while reducing environmental impacts and operational costs. Our findings highlight the potential of IoT and Deep Learning technologies in revolutionizing aquaculture practices, promoting sustainability, and enhancing food security in fisheries

关键词： Deep learning Temperature sensors Productivity Schedules Fisheries Real-time systems Feeds Sustainable development Remote monitoring Water resources

来源：评论

学校读者我要写书评

暂无评论

PROJECTION METHOD FOR QUASIPERIODIC ELLIPTIC EQUATIONS AND APPLICATION TO QUASIPERIODIC HOMOGENIZATION

arXiv

引用

arXiv 2024年

作者： Jiang, Kai Li, Meng Zhang, Juan Zhang, Lei Hunan Key Laboratory for Computation and Simulation in Science and Engineering Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education School of Mathematics and Computational Science Xiangtan University Hunan Xiangtan411105 China School of Mathematical Sciences Institute of Natural Sciences MOE-LSC Shanghai Jiao Tong University Shanghai200240 China

In this study, we address the challenge of solving elliptic equations with quasiperiodic coefficients. To achieve accurate and efficient computation, we introduce the projection method, which enables the embedding of quasiperiodic systems into higher-dimensional periodic systems. To enhance the computational efficiency, we propose a compressed storage strategy for the stiffness matrix by its multi-level block circulant structure, significantly reducing memory requirements. Furthermore, we design a diagonal preconditioner to efficiently solve the resulting high-dimensional linear system by reducing the condition number of the stiffness matrix. These techniques collectively contribute to the computational effectiveness of our proposed approach. Convergence analysis shows the spectral accuracy of the proposed method. We demonstrate the effectiveness and accuracy of our approach through a series of numerical examples. Moreover, we apply our method to achieve a highly accurate computation of the homogenized coefficients for a quasiperiodic multiscale elliptic *** Codes 65D05, 65D15, 65F08, 35B15, 35B27 Copyright © 2024, The Authors. All rights reserved.

关键词： Stiffness matrix

来源：评论

学校读者我要写书评

暂无评论

Disconnection Flow-Mediated Grain Rotation

arXiv

引用

arXiv 2023年

作者： Qiu, Caihao Salvalaglio, Marco Srolovitz, David J. Han, Jian Department of Materials Science and Engineering City University of Hong Kong Hong Kong Institute of Scientific Computing TU Dresden Dresden01062 Germany Dresden Center for Computational Materials Science TU Dresden Dresden01062 Germany Department of Mechanical Engineering The University of Hong Kong Pokfulam Road Hong Kong

Interface migration in microstructures is mediated by the motion of line defects with step and dislocation character, i.e., disconnections. We propose a continuum model for arbitrarily-curved grain boundaries or heterophase interfaces accounting for disconnections’ role in grain rotation. Numerical simulations show that their densities evolve as grain size and shape change, generating stresses and increasing or decreasing misorientation;the predictions agree with molecular dynamics simulations for pure capillarity-driven evolution, and are interpreted through an extended Cahn-Taylor model. Copyright © 2023, The Authors. All rights reserved.

关键词： Molecular dynamics

来源：评论

学校读者我要写书评

暂无评论

AN END-TO-END DEEP LEARNING METHOD FOR SOLVING NONLOCAL ALLEN-CAHN AND CAHN-HILLIARD PHASE-FIELD MODELS

arXiv

引用

arXiv 2024年

作者： Geng, Yuwei Burkovska, Olena Ju, Lili Zhang, Guannan Gunzburger, Max Department of Mathematics University of South Carolina United States Computer Science and Mathematics Division Oak Ridge National Laboratory United States Department of Scientific Computing Florida State University United States Oden Institute for Computational Engineering and Sciences University of Texas at Austin United States

We propose an efficient end-to-end deep learning method for solving nonlocal Allen-Cahn (AC) and Cahn-Hilliard (CH) phase-field models. One motivation for this effort emanates from the fact that discretized partial differential equation-based AC or CH phase-field models result in diffuse interfaces between phases, with the only recourse for remediation is to severely refine the spatial grids in the vicinity of the true moving sharp interface whose width is determined by a grid-independent parameter that is substantially larger than the local grid size. In this work, we introduce non-mass conserving nonlocal AC or CH phase-field models with regular, logarithmic, or obstacle double-well potentials. Because of non-locality, some of these models feature totally sharp interfaces separating phases. The discretization of such models can lead to a transition between phases whose width is only a single grid cell wide. Another motivation is to use deep learning approaches to ameliorate the otherwise high cost of solving discretized nonlocal phase-field models. To this end, loss functions of the customized neural networks are defined using the residual of the fully discrete approximations of the AC or CH models, which results from applying a Fourier collocation method and a temporal semi-implicit approximation. To address the long-range interactions in the models, we tailor the architecture of the neural network by incorporating a nonlocal kernel as an input channel to the neural network model. We then provide the results of extensive computational experiments to illustrate the accuracy, structure-preserving properties, predictive capabilities, and cost reductions of the proposed method. Copyright © 2024, The Authors. All rights reserved.

关键词： Partial differential equations

来源：评论

学校读者我要写书评

暂无评论

Convergence analysis of PM-BDF2 method for quasiperiodic parabolic equations

arXiv

引用

arXiv 2024年

作者： Jiang, Kai Li, Meng Zhang, Juan Zhang, Lei Hunan Key Laboratory for Computation and Simulation in Science and Engineering Key Laboratory of Intelligent Computing and Information Processing Ministry of Education Department of Mathematics and Computational Science Xiangtan University Hunan Xiangtan411105 China School of Mathematical Sciences Institute of Natural Sciences MOE-LSC Shanghai Jiao Tong University Shanghai200240 China

Numerically solving parabolic equations with quasiperiodic coefficients is a significant challenge due to the potential formation of space-filling quasiperiodic structures that lack translational symmetry or decay. In this paper, we introduce a highly accurate numerical method for solving time-dependent quasiperiodic parabolic equations. We discretize the spatial variables using the projection method (PM) and the time variable with the second-order backward differentiation formula (BDF2). We provide a complexity analysis for the resulting PM-BDF2 method. Furthermore, we conduct a detailed convergence analysis, demonstrating that the proposed method exhibits spectral accuracy in space and second-order accuracy in time. Numerical results in both one and two dimensions validate these convergence results, highlighting the PMBDF2 method as a highly efficient algorithm for addressing quasiperiodic parabolic equations. Copyright © 2024, The Authors. All rights reserved.

关键词： Convergence of numerical methods

来源：评论

学校读者我要写书评

暂无评论

How to Get Investment Grade Rating in the Age of Explainable Ai?

SSRN

引用

SSRN 2022年

作者： Makwana, Ravi Bhatt, Dhruvil Delwadia, Kirtan Shah, Agam Chaudhury, Bhaskar Group in Computational Science and HPC DA-IICT Gandhinagar India Financial Services Innovation Lab Scheller College of Business Georgia Institute of Technology AtlantaGA United States School of Computational Science & Engineering College of Computing Georgia Institute of Technology AtlantaGA United States

Corporate credit ratings, issued by specialist agencies, provide an assessment about the credit worthiness of a company and acts as a pivotal financial indication to potential investors. It helps provide investors with a concrete idea about the risk associated with the company’s credit investment returns. Every company aims to attain a good credit rating for seeking more investment and lower debt interest rates. Almost all credit rating agencies have a unique rating scale, broadly divided into either investment class or junk class. Since credit rating agencies scrutinize numerous factors to assign rating, it is difficult for a company to construct a simple and comprehensive set of rules for improvising their credit rating. In this paper, we use explainable AI in the form of decision trees to establish an empirical rule on financial ratios to achieve investment class rating. Further, we explore the temporal aspect of the problem by finding the ideal time window for training data. Currently, limited amount of data is openly available for conducting temporal analysis, and hence a primary focus has been laid on creating a large and quality curated dataset. We show that our proposed empirical rule yields a significantly high precision value, offering a valuable and feasible path that companies can follow to gain a better credit rating. © 2022, The Authors. All rights reserved.

关键词： Decision trees

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：