检索结果-内蒙古大学图书馆

Nonnegative Constrained Graph Based Canonical Correlation Analysis for Multi-view Feature Learning

NEURAL procESSING LETTERS 2019年第2期50卷 1215-1240页

作者： Tan, Huibin Zhang, Xiang Lan, Long Huang, Xuhui Luo, Zhigang Natl Univ Def Technol Sci & Technol Parallel & Distributed Proc Changsha 410073 Hunan Peoples R China Natl Univ Def Technol Inst Quantum Informat State Key Lab High Performance Comp Changsha 410073 Hunan Peoples R China Natl Univ Def Technol Coll Comp Changsha 410073 Hunan Peoples R China Natl Univ Def Technol Dept Comp Sci Coll Comp Changsha Hunan Peoples R China

Understanding and analyzing multi-view data is a fundamental research topic of feature learning for a wide range of practical applications such as image classification. Canonical correlation analysis (CCA) is a popular unsupervised method of analyzing multi-view data, which captures common subspace of two groups of variable sets by maximizing the correlations between them. However, traditional CCA ignores the underlying geometric structure within dataset, which shows great power in describing data distribution, and thus fails in some tasks such as classification. To handle this limitation, this paper proposes an improved CCA variant of Nonnegative Constrained Graph regularized CCA (NCGCCA). Specifically, we improve CCA to NCGCCA with the following two contributions. Firstly, we develop a nonnegative constrained graph based self-representation to explore the underlying group-wise structure within dataset. Secondly, based on the proposed informative representation, we offer a graph embedding schema to incorporate the underlying structure into CCA. Experiments of image classification on four face datasets including Yale, ORL, UMIST, and YaleB demonstrate the efficacy of NCGCCA compared with existing baseline CCA methods.

关键词： Canonical correlation analysis Face recognition Sparse projection Cross-view correlation

来源：评论

学校读者我要写书评

暂无评论

Density-Attentive Head Detector for Crowd Counting 3

Density-Attentive Head Detector for Crowd Counting

引用

3rd International Conference on Data Science and Business Analytics (ICDSBA)

作者： Zhao, Ziyu Shen, Xingyu Lan, Long Yin, Xiaoyao Zhang, Xiang Che, Yonggang Luo, Zhigang Natl Univ Def Technol Coll Comp Changsha Hunan Peoples R China Natl Univ Def Technol Inst Quantum Informat Changsha Peoples R China Natl Univ Def Technol State Key Lab High Performance Comp Changsha Peoples R China Natl Univ Def Technol Sci & Technol Parallel & Distributed Proc Changsha Peoples R China

ISBN: (纸本)9781728146447

In crowd counting, regression-based method shows better performance in extreme density scenes by introduces a density map. However, the regression-based method fails to locate the positions of each head, which significantly restricts its applications. Detection-based method counts each head with their accurate locations but works only in middle or low-level density scenes. In this paper, a joint learning method, named Density-attentive Head Detector (DAHD) is developed to overcome their respective shortcomings via a multi-task training procedure. Specifically, to guarantee the sensitivity of the detector to small or partially occluded heads, we carefully equip the detector with a density map which learned from regression module. Moreover, a novel Dilated Feature Pyramid Network (DFPN) is introduced to our method to enlarge the receptive field of convolutional kernel, bringing confirmative additional benefits to identify small heads. Experiments on the popular ShanghaiTech and Mall datasets confirm the improved performance of DAHD compared with the current detection-based approaches, and a comparable performance to regression-based approaches in term of counting.

关键词： head detector crowd counting attention map

来源：评论

学校读者我要写书评

暂无评论

Multiple CNN-based Tasks Scheduling across Shared GPU Platform in Research and Development Scenarios 20

Multiple CNN-based Tasks Scheduling across Shared GPU Platfo...

引用

20th IEEE International Conference on High Performance Computing and Communications (HPCC) / 16th IEEE International Conference on Smart City (SmartCity) / 4th IEEE International Conference on Data Science and Systems (DSS)

作者： Chen, Zhaoyun Luo, Lei Quan, Wei Shi, Yang Yu, Jie Wen, Mei Zhang, Chunyuan Natl Univ Def Technol Coll Comp Changsha Hunan Peoples R China Natl Key Lab Parallel & Distributed Proc Changsha Hunan Peoples R China

ISBN: (纸本)9781538666142

In the scope of numerous AI enterprises and research institutes, a shared server or cluster, which are based on commodity GPU hardwares, need to process multiple diverse CNN-based tasks simultaneously which are submitted by different developers and researchers. Scheduling and processing multiple CNN-based tasks, including training and batch inference, are a significant challenge in these practical scenarios. Previous studies, which focus on either the latency of a single training task or the throughput of multiple inference tasks, cannot effectively exploit the limited system resources available for diverse CNN-based tasks. This paper, for the first time, focuses on this specific AI Research and Development scenario and conducts an series of explorations on characteration and scheduling for CNN-based tasks. In order to evaluate the qualities of processing and scheduling, we propose a series of comprehensive metrics, including user satisfaction and system efficiency. With the metrics, we characterize diverse CNN behaviors of a few typical CNN models under different application and system configurable factors. Then, a heuristic scheduling algorithm informed by our characterization is explored to better allocate computing resources for the upcoming tasks and to schedule them dynamically on the cluster or server. Compared with two baseline strategies, the results, which are evaluated on multi-GPU platforms, show that our proposed algorithm can improve system efficiency by up to 40% and decrease average response latency by around 38% for multiple CNN-based tasks.

关键词： CNN AI Research and Development Scenario Characterizing Scheduling Exploration GPU platform

来源：评论

学校读者我要写书评

暂无评论

Decentralized Multi-Robot Formation Control with Communication Delay and Asynchronous Clock

引用

JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS 2018年第3-4期89卷 465-484页

作者： Peng, Long Guan, Fei Perneel, Luc Fayyad-Kazan, Hasan Timmerman, Martin VUB Dept Elect & Informat Pl Laan 2 B-1050 Brussels Belgium Natl Univ Def Technol Natl Key Lab Parallel & Distributed Proc Changsha 410073 Hunan Peoples R China VUB Brussels Belgium VUB Embedded Syst Lab Brussels Belgium

This paper investigates the leader-follower formation control problem for a group of networked nonholonomic mobile robots that are subject to bounded time-varying communication delays and an asynchronous clock. First we convert the formation control problem into a trajectory tracking problem, and then a fully distributed unified control framework based on the receding horizon control is implemented to converge the tracking errors. By adding an auxiliary acceleration term into the receding horizon controller, the framework is able to solve the impractical velocity jump problem. Considering the time-varying delays, the timing and order features of the messages are utilized to guarantee their logical correctness. To compensate for the delay effect, an improved control framework that exploits the predictability of the receding horizon controller is proposed. The asynchronous clock problem, which makes the communication delay unmeasurable, is studied. We give a definition of the syn point that is inspired from investigation of the property that messages are received out of order in a bounded time-varying delayed network. A novel method that detects the occurrence of syn points is integrated into the control framework to solve the asynchronous clock problem. Finally the effectiveness of the proposed approaches is demonstrated in the Player/Stage simulation environment.

关键词： Leader-follower formation Trajectory tracking Receding horizon control Communication delay Asynchronous clock Syn point

来源：评论

学校读者我要写书评

暂无评论

Enhancing 2D Representation via Adjacent Views for 3D Shape Retrieval 17

Enhancing 2D Representation via Adjacent Views for 3D Shape ...

引用

IEEE/CVF International Conference on Computer Vision (ICCV)

作者： Xu, Cheng Li, Zhaoqun Qiu, Qiang Leng, Biao Jiang, Jingfei Beihang Univ Sch Comp Sci & Engn Beijing Peoples R China Duke Univ Durham NC 27706 USA Beihang Univ Beijing Adv Innovat Ctr Big Data & Brain Comp Beijing Peoples R China Beihang Univ State Key Lab Virtual Real Technol & Syst Beijing Peoples R China Natl Univ Def Technol Natl Lab Parallel & Distributed Proc Changsha Peoples R China

ISBN: (纸本)9781728148038

Multi-view shape descriptors obtained from various 2D images are commonly adopted in 3D shape retrieval. One major challenge is that significant shape information is discarded during 2D view rendering through projection. In this paper, we propose a convolutional neural network based method, Neighbor-Center Enhanced Network, to enhance each 2D view using its neighboring ones. By exploiting cross-view correlations, Neighbor-Center Enhanced Network learns how adjacent views can be maximally incorporated for an enhanced 2D representation to effectively describe shapes. We observe that a very small amount of, e.g., six, enhanced 2D views, are already sufficient for panoramic shape description. Thus, by simply aggregating features from six enhanced 2D views, we arrive at a highly compact yet discriminative shape descriptor. The proposed shape descriptor significantly outperforms state-of-the-art 3D shape retrieval methods on the ModelNet and ShapeNet-Core55 benchmarks, and also exhibits robustness against object occlusion.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Another Robust NMF: Rethinking the Hyperbolic Tangent Function and Locality Constraint

引用

IEEE ACCESS 2019年 7卷 31089-31102页

作者： Shen, Xingyu Zhang, Xiang Lan, Long Liao, Qing Luo, Zhigang Natl Univ Def Technol Sci & Technol Parallel & Distributed Proc Changsha 410073 Hunan Peoples R China Natl Univ Def Technol Coll Comp Changsha 410073 Hunan Peoples R China Natl Univ Def Technol Inst Quantum Informat Changsha 410073 Hunan Peoples R China Natl Univ Def Technol State Key Lab High Performance Comp Changsha 410073 Hunan Peoples R China Harbin Inst Technol Shenzhen Dept Comp Sci & Technol Shenzhen 518000 Peoples R China

Non-negative matrix factorization (NMF) is a classical data analysis tool for clustering tasks. It usually considers the squared loss to measure the reconstruction error, thus it is sensitive to the presence of outliers. Looking into the literature, most of the existing robust NMF models focus on statistics-based robust estimators with known distribution assumptions. Besides those estimators, whether can we seek another function without the distribution assumption to boost the robustness of NMF? To solve the problem, we propose a robust NMF termed as tanh NMF for short, which rethinks the hyperbolic tangent (tanh) function as a robust loss to evaluate the reconstruction error. Moreover, to capture geometric structure within the data, we devise a locality constraint to regularize tanh NMF to model data locality. Owing to the non-convex tanh function, it is non-trivial to optimize tanh NMF. Following the paradigm of the half-quadratic algorithm, we easily solve an adaptive weighted NMF instead of original tanh NMF. The experiments of face clustering on four popular facial datasets with/without corruptions show that the proposed method achieves the satisfactory performance against several representative baselines including NMF and its robust counterparts. This also implies that the proposed tanh function could serve as an alternative robust loss for NMF.

关键词： Non-negative matrix factorization robust NMF the hyperbolic tangent function half-quadratic algorithm locality constraint

来源：评论

学校读者我要写书评

暂无评论

Exploring Efficient NFA Data Structures to Accelerate DFA Generation 2018

Exploring Efficient NFA Data Structures to Accelerate DFA Ge...

引用

2nd International Conference on Cryptography, Security and Privacy (ICCSP)

作者： Xu, Chengcheng Su, Jinshu Chen, Shuhui Natl Univ Def Technol Coll Comp Changsha 410073 Hunan Peoples R China Natl Key Lab Parallel & Distributed Proc Changsha 410073 Hunan Peoples R China

ISBN: (纸本)9781450363617

Deterministic finite automata (DFA) is widely employed in regular expression matching for content-aware applications, such as protocol identification, NIDS, load balancing, traffic billing, etc. Subset construction is the most time-consuming process when converting regular expressions to the corresponding DFA, which brings a great challenge for fast update application scenarios. Subset construction mainly consists of subset computation and subset comparison. The subset comparison is efficiently processed with the trie tree structure, but the subset computation still remains inefficient for the massive and repetitive memory accesses of NFA state transitions. Through the theoretical analysis and the example demonstration, we prove that the inefficiency mainly originates from the traditional list structures for the NFA state transitions. Then, we proposed two improved data structures for the NFA state transitions, namely the ordered list structure and the ordered array structure. These structures leverage the alphabetical order of the transitions to avoid the huge irrelevant memory accesses during subset computation. Experiments on practical rule sets demonstrate that the subset computation procedure achieves a speedup of 6x to 8x and the total DFA generation achieves a speedup of 2x to 4x.

关键词： Regular expression matching DFA generation Subset construction

来源：评论

学校读者我要写书评

暂无评论

Tracking and Characterizing the Competition of Fact Checking and Misinformation: Case Studies

引用

IEEE ACCESS 2018年 6卷 75327-75341页

作者： Shao, Chengcheng Hui, Pik-Mai Cui, Pengshuai Jiang, Xinwen Peng, Yuxing Natl Univ Def Technol Coll Comp Changsha 410073 Hunan Peoples R China Indiana Univ Sch Informat Comp & Engn Bloomington IN 47408 USA Xiangtan Univ MOE Key Lab Intelligent Comp & Informat Proc Xiangtan 411105 Peoples R China Natl Univ Def Technol Natl Key Lab Parallel & Distributed Proc Changsha 410073 Hunan Peoples R China

Massive amounts of misinformation have been spreading over social media during the 2016 U.S. election season, causing wide public concern about our information ecosystem. How does fact checking compete with misinformation for user attentions? How do misinformation and fact checking differ in their spreading patterns? And what strategies are used by social bots to promote the spread of misinformation? In this paper, we address these research questions by analyzing datasets collected by Hoaxy. We answer the first question by conducting: 1) a survival analysis shows that about 70% of claims will be fact-checked in one week and about 800 tweets with claim links are posted during this period, on average and 2) a cross-correlation analysis shows that the sharing of fact-checking articles typically lags that of misinformation by about one day. Regarding the second question, we answer it from three points of view: 1) we find that false claims can become more popular than the corresponding debunking;2) when looking at the distribution of types of tweets, we find that fact checking tends to spread in a conversational way;and 3) bot behavior analysis shows that the most active accounts sharing misinformation behave more like social bots. To better understand how social bots spread misinformation, we conduct case studies to answer the third question. By presenting strategies such as the production of a large number of original tweets, the alternating and hijacking of hashtags, and the injection of content into conversations, we demonstrate how social bots take advantages of the recommendation features of Twitter to amplify the spread of misinformation.

关键词： Twitter data mining social implications of technology misinformation fact checking social bots social media

来源：评论

学校读者我要写书评

暂无评论

Practical privacy-preserving deep packet inspection outsourcing

引用

CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE 2019年第22期31卷

作者： Li, Jie Su, Jinshu Chen, Rongmao Wang, Xiaofeng Chen, Shuhui Natl Univ Def Technol Sch Comp Changsha Hunan Peoples R China Natl Univ Def Technol Natl Key Lab Parallel & Distributed Proc Changsha Hunan Peoples R China

Hardware-based middleboxes are ubiquitous in computer networks, which usually incur high deployment and management expenses. A recently arising trend aims to address those problems by outsourcing the functions of traditional hardware-based middleboxes to high volume servers in a cloud. This technology is promising but still faces a few challenges from different aspects, including privacy concerns, middlebox functionality, and performance. In this paper, we propose two practical approaches to implementing a cloud-based DPI middlebox. The outsourced DPI middlebox performs payload inspection over encrypted traffic while preserving the privacy of both communication data and inspection rules. Our first approach employs a modified reversible sketch structure, which is used for efficient error-free membership testing, and our second approach extends the famous AC pattern matching algorithm to the cipher text domain. We utilize unkeyed one-way hash functions instead of complex cryptographic protocols to achieve the privacy preservation requirements. Our system supports a wide range of real-world inspection rules. We conduct evaluations on the ClamAV rule set, and the experiment results demonstrate the effectiveness of our proposals.

关键词： DPI outsourcing network function virtualization privacy preservation

来源：评论

学校读者我要写书评

暂无评论

NTIRE 2021 NonHomogeneous Dehazing Challenge Report

NTIRE 2021 NonHomogeneous Dehazing Challenge Report

引用

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Ancuti, Codruta O. Ancuti, Cosmin Vasluianu, Florin-Alexandru Timofte, Radu Fu, Minghan Liu, Huan Yu, Yankun Chen, Jun Wang, keyan Chang, Jerome Wang, Xiyao Liu, Jing Xu, Yi Zhang, Xinjian Zhao, Minyi Zhou, Shuigeng Chen, Tianyi Fu, Jiahui Jiang, Wentao Gao, Chen Liu, Si Wang, Yudong Guo, Jichang Li, Chongyi Yan, Qixin Zheng, Sida Zamir, Syed Waqas Arora, Aditya Dudhane, Akshay Khan, Salman Hayat, Munawar Khan, Fahad Shahbaz Shao, Ling Zhang, Haichuan Guo, Tiantong Monga, Vishal Yang, Wenjin Lin, Jin Luo, Xiaotong Huang, Guowen Chen, Shuxin Qu, Yanyun Xu, Kele Yang, Lehan Sun, Pengliang Niu, Xuetong Zheng, Junjun Ruan, Xiaotong Wang, Yunfeng Yang, Jiang Luo, Zhipeng Wang, Sai Xu, Zhenyu Cao, Xiaochun Luo, Jun Zheng, Zhuoran Ren, Wenqi Wang, Tao Chen, Yiqun Leng, Cong Li, Chenghua Cheng, Jian Sung, Chang-Sung Chen, Jun-Cheng Jo, Eunsung Sim, Jae-Young Geethu, M. M. Akhil, K. A. Sreeni, K. G. Jeena, R. S. Zacharias, Joseph Manu, Chippy M. Huang, Zexi Zhang, Baofeng Zhang, Yiwen Li, Jindong Chen, Mianjie Xiao, Quan Su, Qingchao Han, Lihua Huang, Yanting Prajapati, Kalpesh Chudasama, Vishal Patel, Heena Sarvaiya, Anjali Upla, Kishor Raja, Kiran Ramachandra, Raghavendra Busch, Christoph Jing, Hongyuan Huang, Zilong Fu, Yiran Wu, Haoqiang Zha, Quanxing Zhu, Zhiwei Lv, Hejun Univ Politehn Timisoara ETcTI Timisoara Romania UCL ICTEAM Louvain Belgium Swiss Fed Inst Technol Comp Vis Lab Zurich Switzerland McMaster Univ Hamilton ON Canada Xidian Univ Xian Peoples R China Natl Taiwan Univ Taipei Taiwan Univ Chinese Acad Sci Inst Automat Ctr Res Intelligent Syst & Engn Beijing Peoples R China Bilibili Inc Shanghai Peoples R China Fudan Univ Shanghai Peoples R China Beihang Univ Beijing Peoples R China Tianjin Univ Sch Elect & Informat Engn Tianjin 300072 Peoples R China Nanyang Technol Univ NTU Sch Comp Sci & Engn Singapore Singapore Incept Inst Artificial Intelligence IIAI Abu Dhabi U Arab Emirates Penn State Univ Sch Elect Engn & Comp Sci Informat Proc & Algorithms Lab iPAL University Pk PA 16802 USA Xiamen Univ Dept Comp Sci Xiamen Peoples R China Key Lab Parallel & Distributed Proc Changsha Peoples R China Alibaba Inc Shanghai Peoples R China DeepBlue Technol Shanghai Co Ltd Shanghai Peoples R China Chinese Acad Sci Inst Informat Engn State Key Lab Informat Secur Beijing Peoples R China CASIA Beijing Peoples R China CASIA AiRiA Beijing Peoples R China Natl Taiwan Univ Data Sci Degree Program Taipei Taiwan Ulsan Natl Inst Sci & Technol Ulsan South Korea Coll Engn Dept ECE CV Lab Trivandrum Kerala India AICTE Univ Kollam Kerala India South China Univ Techonol Sch Elect & Informat Engn Guangzhou Peoples R China Shenzhen Wave Kingdom Co Ltd Shenzhen Peoples R China Sardar Vallabhbhai Natl Inst Technol Surat India Beijing Union Univ Coll Robot Beijing Peoples R China

ISBN: (纸本)9781665448994

This work reviews the results of the NTIRE 2021 Challenge on Non-Homogeneous Dehazing. The proposed techniques and their results have been evaluated on a novel dataset that extends the NH-Haze datset. It consists of additional 35 pairs of real haze free and nonhomogeneous hazy images recorded outdoor. The nonhomogeneous haze has been introduced in the outdoor scenes by using a a professional setup that imitates the real conditions of haze scenes. 327 participants registered in the challenge and 23 teams competed in the final testing phase. The proposed solutions gauge the state-of-the-art in image dehazing.

关键词： Computer vision Correlation Conferences Architecture Computer architecture Pattern recognition Testing

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：