检索结果-内蒙古大学图书馆

LSSLP – Local structure sensitive label propagation

information Sciences 2016年 332卷 19-32页

作者： Zhenfeng Zhu Jian Cheng Yao Zhao Jieping Ye Institute of Information Science Beijing Jiaotong University Beijing 100044 China Beijing Key Laboratory of Advanced Information Science and Network Technology Beijing 100044 China National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences (CAS) 100190 China Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor MI 48109-2218 USA

Label propagation is an approach to iteratively spread the prior state of label confidence associated with each of samples to its neighbors until achieving a global convergence state. Such process has been shown to hold close connection with a general graph-based regularization framework. Within this framework, a closed- form linear system can be built to carry out label propagation. In this paper, to address several issues inherent with previous graph-based label propagation framework, we propose a reformulated one, i.e., local structure sensitive label propagation ( LSSLP ). By associating each graph vertex with a local structure sensitive tuning factor, the empirical loss error on each vertex can be controlled preferably to keep consistent with the commonly preconditioned ‘ cluster assumption ’ of data structure. Out of consideration for information conservation, we relax the state conservation constraint of label confidence from labeled samples proposed by Belkin et al. (2004) to a more general form. Meanwhile, an inverse-warping procedure is incorporated into the proposed local structure sensitive label propagation framework to maintain large and stable enough classification margin. Based on the felicitous inversion technique for blocked matrix, we extend LSSLP to its incremental and inductive versions and also present computationally efficient implementation of it. Experimental results demonstrate the performance of the reformulated regularization framework for label propagation is much competitive.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Leveraging Implicit Relative Labeling-Importance information for Effective Multi-label Learning

Leveraging Implicit Relative Labeling-Importance Information...

引用

IEEE International Conference on Data Mining (ICDM)

作者： Yu-Kun Li Min-Ling Zhang Xin Geng Key Laboratory of Computer Network and Information Integration (Southeast University) Ministry of Education China

ISBN: (纸本)9781467395052

In multi-label learning, each training example is represented by a single instance while associated with multiple labels, and the task is to predict a set of relevant labels for the unseen instance. Existing approaches learn from multi-label data by assuming equal labeling-importance, i.e. all the associated labels are regarded to be relevant while their relative importance for the training example are not differentiated. Nonetheless, this assumption fails to reflect the fact that the importance degree of each associated label is generally different, though the importance information is not explicitly accessible from the training examples. In this paper, we show that effective multi-label learning can be achieved by leveraging the implicit relative labeling-importance (RLI) information. Specifically, RLI degrees are formalized as multinomial distribution over the label space, which are estimated by adapting an iterative label propagation procedure. After that, the multi-label prediction model is learned by fitting the estimated multinomial distribution as regularized with popular multi-label empirical loss. Comprehensive experiments clearly validate the usefulness of leveraging implicit RLI information to learn from multi-label data.

关键词： Yttrium Training Reliability Predictive models Symmetric matrices Semantics Estimation

来源：评论

学校读者我要写书评

暂无评论

Comparison of Various Neural network Language Models in Speech Recognition

Comparison of Various Neural Network Language Models in Spee...

引用

International Conference on information Science and Control Engineering (ICISCE)

作者： Lingyun Zuo Xin Wan Jian Liu IACAS Key Laboratory of Speech Acoustics and Content Beijing China XTIPC CAS Xinjiang Laboratory of Minority Speech and Language Information Processing Beijing China National Computer network Emergency Response technical Team Coordination Center Beijing China

ISBN: (纸本)9781509025367

In recent years, research on language modeling for speech recognition has increasingly focused on the application of neural networks. However, the performance of neural network language models strongly depends on their architectural structure. Three competing concepts have been developed: Firstly, feed forward neural networks representing an n-gram approach, Secondly, recurrent neural networks that may learn context dependencies spanning more than a fixed number of predecessor words, Thirdly, the long short-term memory (LSTM) neural networks can fully exploits the correlation on a telephone conversation corpus. In this paper, we compare count models to feed forward, recurrent, and LSTM neural network in conversational telephone speech recognition tasks. Furthermore, we put forward a language model estimation method introduced the information of history sentences. We evaluate the models in terms of perplexity and word error rate, experimentally validating the strong correlation of the two quantities, which we find to hold regardless of the underlying type of the language model. The experimental results show that the performance of LSTM neural network language model is optimal in n-best lists re-score. Compared to the first pass decoding, the relative decline in average word error rate is 4.3% when using ten candidate results to re-score in conversational telephone speech recognition tasks.

关键词： Speech recognition Error analysis Computational modeling Logic gates Recurrent neural networks Speech

来源：评论

学校读者我要写书评

暂无评论

An improved community discovery algorithm in weighted social networks

引用

ICIC Express Letters 2016年第1期10卷 35-41页

作者： Guo, Jingfeng Liu, Miaomiao Liu, Linlin Chen, Xiao College of Information Science and Engineering Yanshan University No. 438 West Hebei Ave Qinhuangdao China College of Information Science and Engineering Northeast Petroleum University No. 199 Fazhan Road Daqing China The Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province No. 438 West Hebei Ave Qinhuangdao China

The algorithm AGMA (Automatic Graph Mining Algorithm) is improved and a novel community discovery algorithm, namely CRMA (Clustering Re-clustering Merging Algorithm) is proposed which can realize more reasonable community division for weighted social networks. Firstly, in the re-clustering stage, AGMA neglected the clustering situation of neighbors of current node and it did not take the weights of edges connecting the current node and its unclustered neighbors into account, which led to that some nodes cannot be clustered or their clustering was unreasonable. Aiming to this, the concept of the connection compactness between the node and the community is introduced, and then through clustering and re-clustering each node would be divided into the community that has the largest connection compactness with the node according to the node weight, the edge weight and the friend coecient between the node and the community. Secondly, for community division in some unweighted networks and signed networks, if there are many nodes whose clustered neighbors are less than its unclustered neighbors, the division results of AGMA tend to result in the lower modularity. In view of this problem, the concepts of positive edge weight density, the cluster density and the connection coecient between clusters are proposed. Lastly, communities are merged based on the above three concepts after re-clustering and the modularity is eectively enhanced. The higher correctness and better generality of the improved algorithm are veried through experiments. © 2016 ICIC International.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Some classes of complete permutation polynomials over Fq

引用

Science China Mathematics 2015年第10期58卷 2081-2094页

作者： WU GaoFei LI Nian HELLESETH Tor ZHANG YuQing State Key Laboratory of Integrated Service Networks Xidian University Information Security and National Computing Grid Laboratory Southwest Jiaotong University Department of Informatics University of Bergen National Computer Network Intrusion Protection Center University of Chinese Academy of Sciences

By using a powerful criterion for permutation polynomials, we give several classes of complete permutation polynomials over finite fields. First, two classes of complete permutation monomials whose exponents are of Niho type are presented. Second, for any odd prime p, we give a sufficient and necessary condition for a-1xdto be a complete permutation polynomial over Fp4 k, where d =(p4k-1)/(pk-1)+ 1 and a ∈ F*p4k. Finally, we present a class of complete permutation multinomials, which is a generalization of recent work.

关键词： finite field complete permutation polynomials Walsh transform Niho exponents Dickson polynomials

来源：评论

学校读者我要写书评

暂无评论

Skin panel design for a morphing mechanism

引用

ICIC Express Letters, Part B: Applications 2016年第10期7卷 2161-2167页

作者： Zhao, Qiang Guo, Xliuan Xi, Fengfeng Key Laboratory for Computer Virtual Technology and System Integration of Hebei Province College of Information Science and Engineering Yanshan University No. 438 Hebei Avenue Qinhuangdao066004 China Department of Aerospace Engineering Ryerson University TorontoONM5B 2K3 Canada

A segmentation method is proposed to aid rigid skin panel design for an existing modular morphing wing mechanism. The method divides the surface of the modular morphing mechanism into panels through geometry discretization, and then the skin which is made up of these rigid panels will have better rigidity and it will morph more free without any gap. The method comprises three basic morphing types based on the motion of upper platform. Wherein, the mechanism side is firstly divided into a plurality of panels and each of them is approximated to flat, and then shape changes of one panel before and after morphing are analyzed. Therefore, panels can be designed as moveable and fixed patches to achieve seamless morphing. At last, computer simulation is set up to further verify the correctness of the method. © 2016 ICIC International.

关键词： Morphing mechanism Seamless morphing Skin panel Surface segmentation

来源：评论

学校读者我要写书评

暂无评论

Trusted Boolean Search on Cloud Using Searchable Symmetric Encryption

Trusted Boolean Search on Cloud Using Searchable Symmetric E...

引用

IEEE International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom)

作者： Cong Zuo James Macindoe Siyin Yang Ron Steinfeld Joseph K. Liu School of Computer and Information Engineering Zhejiang Gongshang University Zhejiang P.R. China Faculty of Information Technology Clayton Campus Monash University Australia Fujian Provincial Key Laboratory of Network Security and Cryptology Fujian Normal University Fuzhou Fujian China

ISBN: (纸本)9781509032068

A Searchable Symmetric Encryption (SSE) scheme allows a server to search a user's data without having to decrypt the data. This provides the user with a high degree of privacy and is particularly useful when data is stored on Cloud. Numerous SSE schemes have already been proposed and while most have excellent security properties, few meet high performance requirements and most only support searching for a single keyword at a time. The SSE scheme of Cash et al. (CRYPTO 2013) is notable for its high efficiency on restricted forms of Boolean queries, but has low efficiency for, or does not support, other common forms of boolean queries. In this paper, we propose a generalization and optimization of the Cash et al. SSE scheme, which extends to support a much larger class of Boolean queries, and performs no worse, and often with much higher efficiency than the Cash et al. scheme for the remaining queries.

关键词： Servers Indexes Encryption Complexity theory

来源：评论

学校读者我要写书评

暂无评论

A simulator for swarm AUVs acoustic communication networking 11

A simulator for swarm AUVs acoustic communication networking

引用

11th ACM International Conference on Underwater networks and Systems, WUWNet 2016

作者： Li, Guannan Liu, Jun Wang, Xue Xu, Hongli Cui, Jun-Hong Shenyang Institute of Automation CAS University of Chinese Academy of Sciences No. 114 Nanta Street Shenhe District Shenyang China College of Computer Science and Technology Jilin University State Key Laboratory of Robotics Key Laboratory of System Control and Information Processing Ministry of Education of China Changchun China Shenyang Institute of Automation CAS Northeast University No. 114 Nanta Street Shenhe District Shenyang China Shenyang Institute of Automation CAS No. 114 Nanta Street Shenhe District Shenyang China College of Computer Science and Technology Jilin University Changchun China Underwater Sensor Network Lab. University of Connecticut StorrsCT06269 United States

ISBN: (纸本)9781450346375

This paper presents a simulator for swarm operations designed to verify algorithms for a swarm of autonomous underwater robots (AUVs), specifically for constructing an underwater communication network with AUVs carrying acoustic communication devices. This simulator consists of three nodes: a virtual vehicle node (VV), a virtual environment node (VE), and a visual showing node (VS). The modular design treats AUV models as a combination of virtual equipment. An expert acoustic communication simulator is embedded in this simulator, to simulate scenarios with dynamic acoustic communication nodes. The several simulations we have performed demonstrate that this simulator is easy to use and can be further improved.

关键词： Simulators

来源：评论

学校读者我要写书评

暂无评论

Hyper-compact virtual estimators for big network data based on register sharing

Hyper-compact virtual estimators for big network data based ...

引用

ACM SIGMETRICS International Conference on Measurement and Modeling of computer Systems, SIGMETRICS 2015

作者： Xiao, Qingjun Chen, Shigang Chen, Min Ling, Yibei Department of Computer and Information Science and Engineering University of Florida GainesvilleFL United States Key Lab. of Computer Network and Information Integration Southeast University Ministry of Education China Telcordia Technologies and Applied Research Ericsson United States

ISBN: (纸本)9781450334860

Cardinality estimation over big network data consisting of numerous flows is a fundamental problem with many practical applications. Traditionally the research on this problem focused on using a small amount of memory to estimate each flow's cardinality from a large range (up to 109). However, although the memory needed for each flow has been greatly compressed, when there is an extremely large number of flows, the overall memory demand can still be very high, exceeding the availability under some important scenarios, such as implementing online measurement modules in network processors using only on-chip cache memory. In this paper, instead of allocating a separated data structure (called estimator) for each flow, we take a different path by viewing all the flows together as a whole: Each flow is allocated with a virtual estimator, and these virtual estimators share a common memory space. We discover that sharing at the register (multi-bit) level is superior than sharing at the bit level. We propose a framework of virtual estimators that allows us to apply the idea of sharing to an array of cardinality estimation solutions, achieving far better memory efficiency than the best existing work. Our experiment shows that the new solution can work in a tight memory space of less than 1 bit per flow or even one tenth of a bit per flow-a quest that has never been realized before. © Copyright 2015 ACM.

关键词： Cache memory

来源：评论

学校读者我要写书评

暂无评论

A heuristic scheduling algorithm based on PSO in the cloud computing environment

引用

International Journal of Future Generation Communication and networking 2016年第1期9卷 349-362页

作者： Xue, Shengjun Shi, Wenling Xu, Xiaolong School of Computer and Software Nanjing University of Information Science and Technology Nanjing China Jiangsu Engineering Center of Network Monitoring Nanjing University of Information Science and Technology Nanjing China State Key Laboratory for Novel Software Technology Nanjing University Nanjing China Department of Computer Science and Technology Nanjing University Nanjing China

In the face of a large number of task requests submitted by users, the data center in the cloud need not only to finish these massive tasks but also to satisfy the user's service demand. How to dividethe resources of the system reasonably and schedule these tasks efficiently isa problem that need be solved in the cloud computing. The task scheduling of workflow is a kind of scheduling model for the most partresearched in cloud computing. The scheduling goal is alwaysdecided by user's QoS(Quality of Service)and the scheduling goal of existing scheduling algorithm is alwayssingle. Based on the price model a scheduling algorithm is proposed which can realize the multiple targets in this article, and the scheduling algorithm is Service Cost Optimizationbased on Particle Swarm Optimization(PSO-SC).PSO-SC algorithm can adapt to dynamic cloud environment, and it not only shortens the completion time of tasks but also minimizesuser's cost of task when schedule tasks. In this paper, experiment and analyze the task scheduling of workflow under the cloud computing environment, and the experimental results show that the algorithm proposed in this paper has very good scheduling performance and can reachthe goal of task scheduling of tasks. © 2016 SERSC.

关键词： Scheduling algorithms

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：