检索结果-内蒙古大学图书馆

26th International Conference on Computational Linguistics, COLING 2016

作者： Mou, Lili Song, Yiping Yan, Rui Li, Ge Zhang, Lu Jin, Zhi Key Laboratory of High Confidence Software Technologies MoE China Institute of Software Peking University China Institute of Network Computing and Information Systems Peking Univerity China Institute of Computer Science and Technology Peking University China

ISBN: (纸本)9784879747020

Using neural networks to generate replies in human-computer dialogue systems is attracting increasing attention over the past few years. However, the performance is not satisfactory: the neural network tends to generate safe, universally relevant replies which carry little meaning. In this paper, we propose a content-introducing approach to neural network-based generative dialogue systems. We first use pointwise mutual information (PMI) to predict a noun as a keyword, reflecting the main GIST of the reply. We then propose seq2BF, a "sequence to backward and forward sequences" model, which generates a reply containing the given keyword. Experimental results show that our approach significantly outperforms traditional sequence-to-sequence models in terms of human evaluation and the entropy measure, and that the predicted keyword can appear at an appropriate position in the reply. © 1963-2018 ACL.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Learning to segment with image-level annotations

引用

Pattern Recognition 2016年 59卷 234-244页

作者： Wei, Yunchao Liang, Xiaodan Chen, Yunpeng Jie, Zequn Xiao, Yanhui Zhao, Yao Yan, Shuicheng Institute of information Science Beijing Jiaotong University Beijing 100044 China Beijing Key Laboratory of Advanced Information Science and Network Technology Beijing 100044 China School of Advanced Computing Sun Yat-Sen University Guangzhou 510006 China Vision and Machine Learning Lab National University of Singapore 117583 Singapore People׳s Public Security University of China Beijing 100038 China

Recently, deep convolutional neural networks (DCNNs) have significantly promoted the development of semantic image segmentation. However, previous works on learning the segmentation network often rely on a large number of ground-truths with pixel-level annotations, which usually require considerable human effort. In this paper, we explore a more challenging problem by learning to segment under image-level annotations. Specifically, our framework consists of two components. First, reliable hypotheses based localization maps are generated by incorporating the hypotheses-aware classification and cross-image contextual refinement. Second, the segmentation network can be trained in a supervised manner by these generated localization maps. We explore two network training strategies for achieving good segmentation performance. For the first strategy, a novel multi-label cross-entropy loss is proposed to train the network by directly using multiple localization maps for all classes, where each pixel contributes to each class with different weights. For the second strategy, the rough segmentation mask can be inferred from the localization maps, and then the network is optimized based on the single-label cross-entropy loss with the produced masks. We evaluate our methods on the PASCAL VOC 2012 segmentation benchmark. Extensive experimental results demonstrate the effectiveness of the proposed methods compared with the state-of-the-arts. © 2016 Elsevier Ltd

关键词： Deep learning Semantic segmentation Weakly supervised

来源：评论

学校读者我要写书评

暂无评论

A novel quantum image compression method based on JPEG

arXiv

引用

arXiv 2017年

作者： Wang, Jian Jiang, Nan Zhao, Na School of Computer and Information Technology Beijing Jiaotong University Beijing100044 China Department of Computer Science Purdue University West Lafayette47907 United States Department of Chemistry Purdue University West Lafayette47907 United States College of Computer Science Beijing University of Technology Beijing100124 China Beijing Key Laboratory of Trusted Computing Beijing100124 China National Engineering Laboratory for Critical Technologies of Information Security Classified Protection Beijing100124 China

Quantum image processing has been a hot topic. The first step of it is to store an image into qubits, which is called quantum image preparation. Different quantum image representations may have different preparation methods. In this paper, we use GQIR (the generalized quantum image representation) to represent an image, and try to decrease the operations used in preparation, which is also known as quantum image compression. Our compression scheme is based on JPEG (named from its inventor: the Joint Photographic Experts Group) — the most widely used method for still image compression in classical computers. We input the quantized JPEG coefficients into qubits and then convert them into pixel values. Theoretical analysis and experimental results show that the compression ratio of our scheme is obviously higher than that of the previous compression method. Copyright © 2017, The Authors. All rights reserved.

关键词： Image compression

来源：评论

学校读者我要写书评

暂无评论

Distributed energy balanced routing for wireless sensor network using energy prediction metric

Distributed energy balanced routing for wireless sensor netw...

引用

International Conference on Information Science and Control Engineering

作者： Quan Guo Xiaoying Song Liaoning Key Lab of Network Security and Computing Technology Dalian Neusoft Information Institute College of Computer Science and Engineering Northeastern University

ISBN: (纸本)9781509025367

A considerable amount of energy efficient routing algorithms have been proposed to save energy and prolong network lifetime. Those algorithms mainly focus on forwarding packets along the minimum energy path to the sink to merely minimize energy consumption, which causes an unbalanced distribution of residual energy among sensor nodes, and eventually results in a network partition. In this paper, we propose a distributed energy balanced routing algorithm using energy prediction metric (EBEP). The routing scheme considers not only efficient energy but also energy balance. Firstly, with the help of Markov chain model and state conversion model, the future residual energy of front neighbor nodes can be predicted so that it is the most important local information when a node makes routing decision. Furthermore, the function of cost is adopted by our EBEP routing algorithm in order to synthetically evaluate and then select the best candidate of next step. Finally, the experimental results demonstrate the performance advantages of our EBEP routing algorithm.

关键词： Wireless sensor networks Balancing energy consumption Energy-efficient routing Markov chain model Residual energy

来源：评论

学校读者我要写书评

暂无评论

Joint power allocation and beamforming for non-orthogonal multiple access (NOMA) in 5G millimeter-wave communications

arXiv

引用

arXiv 2017年

作者： Xiao, Zhenyu Zhu, Lipeng Choi, Jinho Xia, Pengfei Xia, Xiang-Gen School of Electronic and Information Engineering Beihang University Beijing100191 China Beijing Key Laboratory for Network-Based Cooperative Air Traffic Management Beijing Laboratory for General Aviation Technology Beijing100191 China Collaborative Innovation Center of Geospatial Technology Wuhan430079 China Korea Republic of School of Electronics and Information Engineering Key Laboratory of Embedded System and Service Computing Tongji University Shanghai200092 China College of Information Engineering Shenzhen University Shenzhen China Department of Electrical and Computer Engineering University of Delaware NewarkDE19716 United States

In this paper we explore non-orthogonal multiple access (NOMA) in millimeter-wave (mmWave) communications (mmWave-NOMA). In particular, we consider a typical problem, i.e., maximization of the sum rate of a 2-user mmWave-NOMA system. In this problem, we need to find the beamforming vector to steer towards the two users simultaneously subject to an analog beamforming structure, while allocating appropriate power to them. As the problem is non-convex and may not be converted to a convex problem with simple manipulations, we propose a suboptimal solution to this problem. The basic idea is to decompose the original joint beamforming and power allocation problem into two sub-problems which are relatively easy to solve: one is a power and beam gain allocation problem, and the other is a beamforming problem under a constant-modulus constraint. Extension of the proposed solution from 2-user mmWave-NOMA to more-user mmWave-NOMA is also discussed. Extensive performance evaluations are conducted to verify the rational of the proposed solution, and the results also show that the proposed sub-optimal solution achieve close-to-bound sum-rate performance, which is significantly better than that of time-division multiple access (TDMA). Copyright © 2017, The Authors. All rights reserved.

关键词： Millimeter waves

来源：评论

学校读者我要写书评

暂无评论

Automatic mobile retinal microaneurysm detection using handheld fundus camera via cloud computing

Automatic mobile retinal microaneurysm detection using handh...

引用

Imaging and Multimedia Analytics in a Web and Mobile World 2016

作者： You, Jane Li, Qin Guo, Zhenhua Department of Computing Hong Kong Polytechnic University Kowloon Hong Kong Shenzhen Institute of Information Technology School of Software Engineering Shenzhen China Shenzhen Key Laboratory of Broaband Network and Multimedia Graduate School at Shenzhen Tsinghua University China

This paper presents a new system to monitor retinal microaneurysm which are regarded as the first sign of diabetic retinopathy(DR). The proposed approach to automatic microaneurysm detection aims to enhance screening large populations. Most of the existing computer-Aided systems for microaneurysm detection are based on the sophisticated medical device in a clinical environment. However, the popular medical devices such as table fundus camera and portable fundus camera are subject to certain limitations for its usage beyond the scope of clinical practice. The challenges include the complexity of operation, cost issue and requirement of professional maintenance, etc. Unlike the conventional approaches, we developed an automatic mobile retinal microaneurysm detection system by using a handheld fundus camera to facilitate retinal healthcare and monitoring with flexibility and convenience. Our system includes: (1) retinal image capturing by handheld fundus camera;(2) retinal image analysis via cloud computing;(3) microaneurysm detection by Multi-orientation Sum of Matched Filter and SVM. The experimental results demonstrate the feasibility of our system by performance improvement on the aspects of speed, accuracy, and convenience. © 2016 Society for Imaging Science and technology.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Classification of Handwriting Number Based on PCANet network with Data Augmentation

Classification of Handwriting Number Based on PCANet Network...

引用

2016 2nd International Conference on Advances in Mechanical Engineering and Industrial Informatics(AMEII 2016)

作者： Tianmei Guo Jiwen Dong Lei Wang Department of Computer Science and Technology Shandong Provincial Key Laboratory of Network based Intelligent Computing University of Jinan

ISBN: (纸本)9781510823808

In this paper we proposed an algorithm for classification of handwriting number which use depth of the structure of the simple convolution neural network-- PCANet network with data *** network with data augmentation consists of four parts,one is data augmentation,two is the basic PCA filters,part three is binary quantization,four is the block-wise *** simple convolution neural network improve some defects of classic convolution neural network such as training time too long,need special tuning parameters and technique *** allows for training networks with many features,making them insensitive to variability between class and *** of experimental results show that the PCANet network with data augmentation algorithms have higher recognition rate than the classification effect with that of SVM algorithm,BP neural network and PCANet.

关键词： handwriting number recognition Deep learning PCANet network data augmentation

来源：评论

学校读者我要写书评

暂无评论

Using Multi-features and Ensemble Learning Method for Imbalanced Malware Classification

Using Multi-features and Ensemble Learning Method for Imbala...

引用

IEEE International Conference on Trust, security and Privacy in computing and Communications (TrustCom)

作者： Yunan Zhang Qingjia Huang Xinjian Ma Zeming Yang Jianguo Jiang Beijing Key Laboratory of Network Security Technology CAS Beijing China Institute of Information Engineering CAS Beijing China

ISBN: (纸本)9781509032068

The ever-growing malware threats in the cyber spacecalls for techniques that are more effective than widely deployedsignature-based detection system. To counter large volumes ofmalware variants, machine learning techniques have been appliedfor automated malware classification. Despite these efforts haveachieved a certain success, the accuracy and efficiency stillremained inadequate to meet demand, especially when thesemachine learning techniques are used in the situation of multipleclass classification and imbalanced training data. Against thisbackdrop, the goal of this paper is to build a malware classificationsystem that could be used to improve the above mentionedsituation. Our system is based on multiple categories of staticfeatures and ensemble learning method. Compared to sometraditional systems it has the following advantages. Firstly, withmultiple categories of features, our system could classify malwareto their corresponding family effectively and efficiently and at thesame time avoid the influence of evasion in certain extent. Ourmethod don't need any unpacking process and extract featuresfrom the bytes file and disassembled asm file directly. Secondly, the system employed two efficient ensemble learning models, namely XGBoost and ExtraTreeClassifer, and also combinedstacking method to construct the final classifier. Finally, weexperimented our system with the dataset provided by Microsofthosted in Kaggle for malware classification competition, andthe final results show that our method could classify malwareto their family effectively and efficiently with the accuracy of0.9972 in training set and logloss of 0.00395 in testing set. Ourwork not only offer insights into how to use multiple features forclassification, but also shed light on how to develop a scalabletechniques for automated malware classification in practice.

关键词： Malware Feature extraction Learning systems Training Stacking Classification algorithms Virtual machining

来源：评论

学校读者我要写书评

暂无评论

A trusted evidence collection method based on the trusted third-party for cloud platform

引用

International Journal of Simulation: Systems, Science and technology 2016年第25期17卷

作者： Wu, Lili Zhan, Jing Zhao, Yong Hu, Jun Li, Ming Beijing Key Laboratory of Trusted Computing Beijing University of Technology Beijing China National Engineering Laboratory for Critical Technologies of Information Security Classified Protection Beijing University of Technology Beijing China The Third Research Institute of Ministry of public security Shanghai China

The characteristics of cloud computing, such as highly centralized resources, dynamic extensibility, and layered service mode, make the trusted evidence collection on the cloud platforms becomes very complex, this article introduces a trusted third party, and proposes a trusted evidence collection method based on the technology of trusted computing. security features provided by TPCM (Trusted Platform Control Module) are used to introduce a CPTECA (Cloud Platform Trusted Evidence Collection Agent) in each layer of cloud platform respectively. The Agent monitors the entity in real-time using TPCM, and objectively collects the trusted evidence on the cloud Platform. Finally, the paper implements a CPTECA functional prototype based on the LSM (Linux security Module). © 2016, UK Simulation Society. All rights reserved.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Frequency Synchronization for Massive MIMO Multi-User Uplink

Frequency Synchronization for Massive MIMO Multi-User Uplink

引用

IEEE Global Communications Conference

作者： Weile Zhang Feifei Gao Hui-Ming Wang MOE Key Lab for Intelligent Networks and Network Security Xi'an Jiaotong University Tsinghua National Laboratory for Information Science and Technology

ISBN: (纸本)9781509013296

In this paper, we propose a new frequency synchronization scheme for multiuser orthogonal frequency division multiplexing (OFDM) uplink with a large-scale uniform linear array (ULA) at base station (BS). Considering that the incident signal at BS from each user can be restricted within a narrow angular spread, the proposed scheme performs carrier frequency offset (CFO) estimation for each user individually through a joint spatial-frequency alignment procedure. The basic idea behind is that with sufficient spatial dimension, the multiuser interference (MUI) effect can be effectively mitigated via beamforming with the steering vectors corresponding to any direction of arrival (DOA) of each user. A multi-branch receive beamforming is further designed for each user which results in an equivalent single user transmission model and the conventional single-user channel estimation and data detection can be carried out. We show that the proposed CFO estimation can be applied efficiently with the aided of fast Fourier transform (FFT). The theoretical CFO estimation performance analysis is also conducted. Various numerical results are provided to verify the proposed studies.

关键词： Frequency synchronization Carrier frequency offset (CFO) estimation Multiuser detection Massive multi-input multi-output (MIMO)

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：