检索结果-内蒙古大学图书馆

arXiv 2019年

作者： Ma, Ke Zeng, Jinshan Xiong, Jiechao Xu, Qianqian Cao, Xiaochun Liu, Wei Yao, Yuan School of Computer Science and Technology University of Chinese Academy of Sciences Beijing100049 China Artificial Intelligence Research Center Peng Cheng Laboratory Shenzhen518055 China School of Computer Information Engineering Jiangxi Normal University NanchangJiangxi330022 China Tencent AI Lab Shenzhen Guangdong China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China State Key Laboratory of Information Security Institute of Information Engineering Chinese Academy of Sciences Beijing100093 China Cyberspace Security Research Center Peng Cheng Laboratory Shenzhen518055 China School of Cyber Security University of Chinese Academy of Sciences Beijing100049 China Department of Mathematics and by courtesy Department of Computer Science and Engineering Hong Kong University of Science and Technology Clear Water Bay Kowloon Hong Kong

—Learning representation from relative similarity comparisons, often called ordinal embedding, gains rising attention in recent years. Most of the existing methods are based on semi-definite programming (SDP), which is generally time-consuming and degrades the scalability, especially confronting large-scale data. To overcome this challenge, we propose a stochastic algorithm called SVRG-SBB, which has the following features: i) achieving good scalability via dropping positive semi-definite (PSD) constraints as serving a fast algorithm, i.e., stochastic variance reduced gradient (SVRG) method, and ii) adaptive learning via introducing a new, adaptive step size called the stabilized Barzilai-Borwein (SBB) step size. Theoretically, under some natural assumptions, we show the O(T1 ) rate of convergence to a stationary point of the proposed algorithm, where T is the number of total iterations. Under the further Polyak-Lojasiewicz assumption, we can show the global linear convergence (i.e., exponentially fast converging to a global optimum) of the proposed algorithm. Numerous simulations and real-world data experiments are conducted to show the effectiveness of the proposed algorithm by comparing with the state-of-the-art methods, notably, much lower computational cost with good prediction performance. Copyright © 2019, The Authors. All rights reserved.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

Stochastic Link Flow Model for Signalized Traffic Networks with Uncertainty in Demand

引用

IFAC-PapersOnLine 2018年第9期51卷 458-463页

作者： Lin, S. Pan, T.L. Lam, W.H.K. Zhong, R.X. Schutter, B. De School of Computer and Control Engineering University of Chinese Academy of Sciences Key Laboratory of System Control and Information Processing Ministry of Education BeijingShanghai China Department of Civil and Structural Engineering The Hong Kong Polytechnic University Hong KongSAR Hong Kong School of Engineering Sun Yat-Sen University Guangzhou China Delft Center for Systems and Control Delft University of Technology Delft Netherlands

In order to investigate the stochastic features in urban traffic dynamics, we propose a Stochastic Link Flow Model (SLFM) for signalized traffic networks with demand uncertainties. In the proposed model, the link traffic state is described using four different link state modes, and the probability for each link state mode is determined based on the stochastic link states. The SLFM model is expressed as a finite mixture approximation of the link state probabilities and the dynamic link flow models for all the four link state modes. Using data from microscopic traffic simulator SUMO, we illustrate that the proposed model can provide a reliable estimation of the link traffic states, and as well as good estimations on the link state uncertainties propagating within a signalized traffic network. © 2018

关键词： Stochastic models Routers Stochastic systems Street traffic control Traffic signals Vehicle actuated signals Demand uncertainty Finite mixtures Stochastic features Stochastic traffic Traffic networks Traffic simulators Uncertainty in demand Urban traffic networks

来源：评论

学校读者我要写书评

暂无评论

Benchmark Computation of Morphological Complexity in the Functionalized Cahn-Hilliard Gradient Flow

arXiv

引用

arXiv 2020年

作者： Christlieb, Andrew Promislow, Keith Tan, Zengqiang Wang, Sulin Wetton, Brian Wise, Steven M. Department of Mathematics Michigan State University East LansingMI48824 United States Department of Computational Mathematics Science and Engineering Michigan State University East LansingMI48824 United States School of Science Wuhan University of Technology Wuhan430070 China School of Mathematics Hunan University Changsha410082 China Hunan Provincial Key Laboratory of Intelligent Information Processing and Applied Mathematics Hunan University Changsha410082 China Department of Mathematics The University of British Columbia VancouverBCV6T 1Z2 Canada Department of Mathematics The University of Tennessee KnoxvilleTN37996 United States

Reductions of the self-consistent mean field theory model of amphiphilic molecules in solvent can lead to a singular family of functionalized Cahn-Hilliard (FCH) energies. We modify these energies, mollifying the singularities to stabilize the computation of the gradient flows and develop a series of benchmark problems that emulate the "morphological complexity" observed in experiments. These benchmarks investigate the delicate balance between the rate of absorption of amphiphilic material onto an interface and a least energy mechanism to disperse the arriving mass. The result is a trichotomy of responses in which two-dimensional interfaces either lengthen by a regularized motion against curvature, undergo pearling bifurcations, or split directly into networks of interfaces. We evaluate a number of schemes that use second order backward differentiation formula (BDF2) type time stepping coupled with Fourier pseudo-spectral spatial discretization. The BDF2-type schemes are either based on a fully implicit time discretization with a preconditioned steepest descent (PSD) nonlinear solver or upon linearly implicit time discretization based on the standard implicit-explicit (IMEX) and the scalar auxiliary variable (SAV) approaches. We add an exponential time differencing (ETD) scheme for comparison purposes. All schemes use a fixed local truncation error target with adaptive time-stepping to achieve the error target. Each scheme requires proper "preconditioning" to achieve robust performance that can enhance efficiency by several orders of magnitude. The nonlinear PSD scheme achieves the smallest global discretization error at fixed local truncation error, however the IMEX and SAV schemes are the most computationally efficient as Copyright © 2020, The Authors. All rights reserved.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

Resource allocation method based on combinatorial double auction mechanism in cloud computing

Resource allocation method based on combinatorial double auc...

引用

IEEE Conference on Industrial Electronics and Applications (ICIEA)

作者： Li Deng Fei Xu Yulin Ren ShengGang Bao Heng He Chao Li College of Computer Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan China

Cloud computing integrates kinds of resources into a huge resource pool and provides different types of resource combination in the resource pool to users in the form of service using virtualization technology. The entities of cloud computing often ignore participants' experience quality and excessive energy consumption in resource trading. A cloud resource allocation method is presented based on power-aware and credible combinatorial double auction (PACCDA). The mechanism comprehensively considers several factors, such as credits of providers and users, and energy costs, to resist dishonesty and malicious behavior in auctions. The simulation results show that, compared with CDARA, PACCDA has better performance on the number of winners, profits of service providers, and profits of consumers.

关键词： Cloud computing Resource management Energy consumption Bandwidth computer science information processing Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Pixels Matching in No Obvious Feature Area in Binocular Vision Based on Peripheral Feature Points

Pixels Matching in No Obvious Feature Area in Binocular Visi...

引用

International Conference on Security, Pattern Analysis, and Cybernetics (SPAC)

作者： Renlong Chen Mingjun Liu Xueyan Gong Jinping Li School of Information Science and Engineering Shandong Provincial Key Laboratory of Network Based Intelligent Computing (University of Jinan) Shandong College and University Key Laboratory of Information Processing and Cognitive Computing in 13th Five-year Jinan China Qilu Institute of Technology Shandong Provincial Key Laboratory of Network Based Intelligent Computing (University of Jinan) Shandong College and University Key Laboratory of Information Processing and Cognitive Computing in 13th Five-year Jinan China

ISBN: (纸本)9781728105529

In binocular vision, the pixel matching of no obvious feature refers to the matching of pixels in the area where the gray value does not change significantly or in the area where there is no significant gradient change in the gray level. The basic idea is that the position of the matching pixels in the area with no obvious features can be determined by the peripheral feature points. First, the camera is calibrated by the chessboard calibration method, and the internal and external parameters of the camera are obtained. The distortion of the left and right images is corrected by the calibration results. Then, the SIFT algorithm is used to extract the feature points, the Euclidean distance threshold is set to determine the better matching points, and the random sample consensus (RANSAC) algorithm is used to eliminate the wrong matching points. Finally, four target matching points with the same distance and direction information are obtained, and the final target matching points are obtained by averaging the four target matching points. Experiments show that any pair of pixels in the unmarked area can be matched accurately.

关键词： Binocular vision Pixel matching with no obvious feature area SIFT RANSAC Peripheral feature points Binocular vision pixel Feature Points Sifting gray value Cameras

来源：评论

学校读者我要写书评

暂无评论

DABC-Net for robust pneumonia segmentation and prediction of COVID-19 progression on chest CT scans

Research Square

引用

Research Square 2020年

作者： Zhang, Xiao-Yong Yu, Ziqi Han, Xiaoyang Zhao, Botao Zhuo, Yaoyao Ren, Yan Xue, Xiangyang Lamm, Lorenz Feng, Jianfeng Marr, Carsten Shan, Fei Peng, Tingying Institute of Science and Technology for Brain-Inspired Intelligence Fudan University Shanghai200433 China Key Laboratory of Computational Neuroscience and Brain-Inspired Intelligence Ministry of Education Shanghai200433 China Department of Radiology Shanghai Public Health Clinical Center Fudan University Shanghai201508 China Department of Radiology Huashan Hospital Fudan University Shanghai200433 China Shanghai Key Laboratory of Intelligent Information Processing School of Computer Science Fudan University Shanghai200433 China Institute of Computational Biology Helmholtz Zentrum München Ingolstädter Landstraße 1 D-85764 Germany Helmholtz AI Helmholtz Zentrum München Ingolstädter Landstraße 1 NeuherbergD-85764 Germany

Currently, reliable, robust and ready-to-use CT-based tools for prediction of COVID-19 progression are still lacking. To address this problem, we present DABC-Net, a novel deep learning (DL) tool that combines a 2D U-net for intra-slice spatial information processing, and a recurrent LSTM network to leverage inter-slice context, for automatic volumetric segmentation of lung and pneumonia lesions. We evaluate DABC-Net on more than 10,000 radiologists-labeled CT slices from four different cohorts. Compared to state-of-the-art segmentation tools, DABC-Net is much faster, more robust, and able to estimate segmentation uncertainty. Based only on the first two CT scans within 3 days after admission from 656 longitudinal CT scans, the AUC of our DBAC-Net for disease progression prediction reaches 93%. We release our tool as a GUI for patient-specific prediction of pneumonia progression, to provide clinicians with additional assistance to triage patients at early days after the diagnosis and to optimize the assignment of limited medical resources, which is of particular importance in current critical COVID-19 pandemic. © 2020, CC BY.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Reform and Construction of computer Major on Local Colleges under the Background of Emerging Engineering Education

Reform and Construction of Computer Major on Local Colleges ...

引用

3rd International Conference on Politics,Economics and Law(ICPEL 2018)

作者： Yaojie Chen Xin Yuan Hai Zhou Zhijing Wan School of Computer Science and Technology Wuhan university of Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System

In recent years, with the accelerated development of the global new economy, the country desperately need the support of new engineering talents, which raises higher requirements for the development of higher education in *** in 2016, the "emerging engineering education" was first proposed in China, which is more significant on colleges across the country. computer major that is an enormous engineering category has trained a large number of talents for the ***, how to pay attention to the new needs of social development and cultivate more high-quality computer professionals is the key for computer major to take the road of connotation development in the future. In this paper, under the background of emerging engineering education, how to carry out teaching reform in local colleges, how to promote the cross-integration of computer subject, and how to strengthen the integration of research, education and industry are discussed in depth. Therefore, the construction of a new project education concept is the future core direction for local colleges.

关键词： emerging engineering education computer major local colleges interdisciplinary innovation

来源：评论

学校读者我要写书评

暂无评论

Neural aesthetic image reviewer

arXiv

引用

arXiv 2018年

作者： Wang, Wenshan Yang, Su Zhang, Weishan Zhang, Jiulong Shanghai Key Laboratory of Intelligent Information Processing School of Computer Science Fudan University China University of Petroleum Xi'an University of Technology

Recently, there is a rising interest in perceiving image aesthetics. The existing works deal with image aesthetics as a classification or regression problem. To extend the cognition from rating to reasoning, a deeper understanding of aesthetics should be based on revealing why a high- or low-aesthetic score should be assigned to an image. From such a point of view, we propose a model referred to as Neural Aesthetic Image Reviewer, which can not only give an aesthetic score for an image, but also generate a textual description explaining why the image leads to a plausible rating score. Specifically, we propose two multi-task architectures based on shared aesthetically semantic layers and task-specific embedding layers at a high level for performance improvement on different tasks. To facilitate researches on this problem, we collect the AVA-Reviews dataset, which contains 52,118 images and 312,708 comments in total. Through multi-task learning, the proposed models can rate aesthetic images as well as produce comments in an end-to-end manner. It is confirmed that the proposed models outperform the baselines according to the performance evaluation on the AVA-Reviews dataset. Moreover, we demonstrate experimentally that our model can generate textual reviews related to aesthetics, which are consistent with human perception. Copyright © 2018, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Application of Deep Learning in Software Security Detection

Application of Deep Learning in Software Security Detection

引用

2018 International Conference on Computational Science and Engineering(ICCSE 2018)

作者： Lin Li Ying Ding Jiacheng Mao College of Computer Science and Technology Wuhan University of Science and Technology Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System

Machine learning is knowledge of learning rules from data through computational models and algorithms. It has been applied in various fields that require mining rules from complex data, and has become one of the most core technologies in the field of artificial intelligence in the broad sense. In recent years, a variety of deep neural networks have made remarkable achievements in a large number of machine learning problems, forming a new branch of machine learning, deep learning, and also raising a new climax of machine learning theory, method and application research. Software security is a highly concerned issue at present. Vulnerability detection technology is an important measure to improve software security. In order to make software application more secure, it is necessary to discuss the application of deep learning in software security detection.

关键词： Deep learning Software security Detection RNN

来源：评论

学校读者我要写书评

暂无评论

Controlling Expressivity using Input Codes in Neural Network based TTS

Controlling Expressivity using Input Codes in Neural Network...

引用

Asian Conference on Affective Computing and Intelligent Interaction (ACII Asia)

作者： Xiaolian Zhu Lei Xie Xiao Chen Xiaoyan Lou Xuan Zhu Xingjun Tan Shaanxi Provincial Key Laboratory of Speech and Image Information Processing School of Computer Science Northwestern Polytechnical University Xi’an Hebei University of Economics and Business Shijiazhuang China Shaanxi Provincial Key Laboratory of Speech and Image Information Processing School of Computer Science Northwestern Polytechnical University Xi’an China Language Computing Lab Samsung R&D Institute of China Beijing China

ISBN: (纸本)9781538653128

This paper presents a study on the use of input codes in the neural network acoustic modeling for expressive TTS. Specifically, we use different kinds of input codes, augmented with the linguistic features, as the input of a BLSTM-based acoustic model, to control the expressivity of the synthesized speech. The input codes, in one-hot representation, include dialogue code, sentiment code and sentence position code. The dialogue code indicates whether the text is a dialogue or narration in an audiobook story. The sentiment code is obtained from a sentiment analysis tool, which labels each sentence as positive, negative and neutral. The sentence position code indicates the position of the sentence in the paragraph. We believe these codes are highly related to the expressiveness of the audiobook speech. Experiments on the data from the Blizzard Challenge 2017 demonstrate the effectiveness of the use of input codes in the neural network approach for expressive TTS.

关键词： Neural networks Linguistics Acoustics Hidden Markov models Speech synthesis Adaptation models Speech coding

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：