检索结果-内蒙古大学图书馆

6th International Conference on Learning Representations, ICLR 2018

作者： Lau, Tim Tsz-Kit Zeng, Jinshan Wu, Baoyuan Yao, Yuan Department of Mathematics Hong Kong University of Science and Technology Hong Kong School of Computer and Information Engineering Jiangxi Normal University China Tencent AI Lab

Training deep neural networks (DNNs) efficiently is a challenge due to the associated highly nonconvex optimization. The backpropagation (backprop) algorithm has long been the most widely used algorithm for gradient computation of parameters of DNNs and is used along with gradient descent-type algorithms for this optimization task. Recent work have shown the efficiency of block coordinate descent (BCD) type methods empirically for training DNNs. In view of this, we propose a novel algorithm based on the BCD method for training DNNs and provide its global convergence results built upon the powerful framework of the Kurdyka-Lojasiewicz (KL) property. Numerical experiments on standard datasets demonstrate its competitive efficiency against standard optimizers with backprop. © 6th International Conference on Learning Representations, ICLR 2018 - Workshop Track Proceedings. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Four-dimensional polarisation-ring-switching for dispersion-managed optical fibre systems

arXiv

引用

arXiv 2019年

作者： Chen, Bin Okonkwo, Chigo Hafermann, Hartmut Alvarado, Alex Department of Electrical Engineering Eindhoven University of Technology Netherlands School of Computer and Information Hefei University of Technology China Mathematical and Algorithmic Sciences Lab Paris Research Center Huawei Technologies France SASU France

The recently introduced 4D 64-ary polarisation-ring-switching format is investigated in dispersion-managed systems. Numerical simulations show a reach increase of 25% with respect to PM-8QAM. This gain is achieved from the nonlinear tolerance of the format and a 4D demapper using correlated noise assumptions. Copyright © 2019, The Authors. All rights reserved.

关键词： Polarization

来源：评论

学校读者我要写书评

暂无评论

On security improvement of adaptive pixel pair matching with modified searching mechanism

引用

International Journal of Network Security 2019年第2期21卷 230-235页

作者： Hong, Wien Zheng, Shuozhen Zhou, Xiaoyu School of Electrical and Computer Engineering Nanfang College of Sun Yat-Sen University 882 Wenquan Rd. Conghua district Guangzhou510970 China School of Computer and Software Nanjing University of Information Science and Technology 219 Niliu Rd. Nanjing Jiangsu210044 China

This paper proposes a data hiding scheme that improves the adaptive pixel pair matching (APPM) method. Based on pixel pair matching, APPM employs a pixel pair as an embedding unit, and uses a specially designed reference table that minimizing the embedding distortion for data embedment. Although APPM has the capability to em bed secret digit in any notational system, it is vulnerable to the detection by the RS scheme if digits in 4-ary no tational system are embedded into images with large at area (pixels having the similar grayscale values) such car toon images. A modified version of APPM is proposed in this paper by using a revised pixel pair replacement mechanism (PPRM). With the proposed PPRM method, the stego image not only is totally undetectable by the RS scheme but also provides the equivalent image quality of the original APPM method. © 2019 Femto Technique Co., Ltd.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

引用

25th International Conference on Neural information Processing, ICONIP 2018

作者： Liu, Wenfeng Liu, Peiyu Yi, Jing Yang, Yuzhen Liu, Weitong Li, Nana School of Information Science and Engineering Shandong Normal University Jinan250014 China School of Computer Heze University Heze274015 China Shandong Provincial Key Laboratory for Distributed Computer Software Novel Technology Jinan250014 China School of Computer Science and Technology Shandong Jianzhu University Jinan250101 China

ISBN: (纸本)9783030041816

How to effectively measure the similarity between two sentences is a challenging task in natural language processing. In this paper, we propose a sentence similarity comparison method that combines word embeddings and syntactic structure. First of all, by generating the corresponding syntactic tree, we synthetically analyze the two sentences and block them according to the syntactic components. Secondly, we prune the syntactic tree, remove the stop words and perform morphological restoration. Then, some important operations will be performed, such as passive flipping, negative flipping, and so on. Finally, the similarity of two sentence pairs is calculated by weighting the block embeddings of the syntactic tree. Experiments show the effectiveness of this method. © Springer Nature Switzerland AG 2018.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

Four-dimensional polarisation-ring-switching for dispersion-managed optical fibre systems

Four-dimensional polarisation-ring-switching for dispersion-...

引用

45th European Conference on Optical Communication (ECOC 2019)

作者： Bin Chen Chigo Okonkwo Hartmut Hafermann Alex Alvarado Department of Electrical Engineering Eindhoven University of Technology The Netherlands School of Computer and Information Hefei University of Technology China Mathematical and Algorithmic Sciences Lab Paris Research Center Huawei Technologies France SASU France

ISBN: (纸本)9781839531859

关键词： CODED MODULATION GENERALISED MUTUAL information DISPERSION MANAGED SYSTEM

来源：评论

学校读者我要写书评

暂无评论

Undersampling Near Decision Boundary for Imbalance Problems

Undersampling Near Decision Boundary for Imbalance Problems

引用

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Jianjun Zhang Ting Wang Wing W. Y. Ng Shuai Zhang Chris D. Nugent Guangdong Provincial Key Lab of Computational Intelligence and Cyberspace Information School of Computer Science and Engineering South China University of Technology Guangzhou China School of Computing Ulster University Jordanstown Belfast United Kingdom

ISBN: (纸本)9781728128177

Undersampling the dataset to rebalance the class distribution is effective to handle class imbalance problems. However, randomly removing majority examples via a uniform distribution may lead to unnecessary information loss. This would result in performance deterioration of classifiers trained using this rebalanced dataset. On the other hand, examples have different sensitivities with respect to class imbalance. Higher sensitivity means that this example is more easily to be affected by class imbalance, which can be used to guide the selection of examples to rebalance the class distribution and to boost the classifier performance. Therefore, in this paper, we propose a novel undersampling method, the UnderSampling using Sensitivity (USS), based on sensitivity of each majority example. Examples with low sensitivities are noisy or safe examples while examples with high sensitivities are borderline examples. In USS, majority examples with higher sensitivities are more likely to be selected. Experiments on 20 datasets confirm the superiority of the USS against one baseline method and five resampling methods.

关键词： Sensitivity Undersampling Imbalance learning USS Undersampling HIGH SENSITIVITY information loss Balance impairment IMBALANCE rebalance Classifiers antimicrobial susceptibility NATIONAL EMERGENCY ALARM REPEATER Dataset

来源：评论

学校读者我要写书评

暂无评论

A cyber-enabled mission-critical system for post-flood response: Exploiting TV white space as network backhaul links

A cyber-enabled mission-critical system for post-flood respo...

引用

作者： Rahman, M.D. Arafatur Asyhari, A. Taufiq Azad, Saiful Hasan, M.D. Munirul Munaiseche, Cindy P.C. Krisnanda, Made Faculty of Computer Systems and Software Engineering Universiti Malaysia Pahang Kuantan26300 Malaysia IBM Center of Excellence and Earth Resources and Sustainability Center Universiti Malaysia Pahang Kuantan26300 Malaysia School of Computing Electronics and Mathematics Coventry University CoventryCV1 5FB United Kingdom Telkom University Jalan Telekomunikasi Terusan Buah Batu Bandung40257 Indonesia Faculty of Engineering Department of Informatics Universitas Negeri Manado Tondano95618 Indonesia Faculty of Engineering Department of Information and Communication Technology Education Universitas Negeri Manado Tondano95618 Indonesia

A crucial problem in post-flood recovery actions is the ability to rapidly establish communication and collaboration among rescuers to conduct timely and effective search and rescue (SAR) mission given disrupted telecommunication infrastructure to support the service. Aimed at providing such proximity service (ProSe) for mission-critical data exchange in the post-flood environment, the majority of existing solutions rely heavily upon ad-hoc networking approaches, which suffer from restricted communication range and the limited scope of interaction. As an effort to broaden the ProSe coverage and expand integrated global-local information exchange in the post-flood SAR activities, this paper proposes a novel network architecture in the form of a cyber-enabled mission-critical system (CEMCS) for acquiring and communicating post-flood emergency data by exploiting TV white space spectrum as network backhaul links. The primary method of developing the proposed system builds upon a layered architecture of wireless local, regional and wide-area communications, and incorporates collaborative network components among these layers. The desirable functionalities of CEMCS are showcased through formulation and the development of an efficient global search strategy exploiting a wide range of collaboration among network agents. The simulation results demonstrate the capability of CEMCS to provide ProSe in the post-flood scenarios as reflected by reliable network performance (e.g., packet delivery ratio nearing 80%–90%) and the optimality of efficient search algorithm. © 2017 Oxford University Press. All rights reserved.

关键词： Cyber Physical System

来源：评论

学校读者我要写书评

暂无评论

MUSEFood: Multi-Sensor-Based Food Volume Estimation on Smartphones

MUSEFood: Multi-Sensor-Based Food Volume Estimation on Smart...

引用

Symposia and Workshops on Ubiquitous, Autonomic and Trusted Computing, UIC-ATC

作者： Junyi Gao Weihao Tan Liantao Ma Yasha Wang Wen Tang Key Laboratory of High Confidence Software Technologies Ministry of Education Beijing China Peking University Information Technology Institute (Tianjin Binhai) Tianjin China School of Electronics Engineering and Computer Science Peking University Beijing China National Engineering Research Center for Software Engineering Peking University Beijing China Department of Nephrology Peking University Third Hospital Beijing China

ISBN: (数字)9781728140346

ISBN: (纸本)9781728140353

Researches have shown that diet recording can help people increase awareness of food intake and improve nutrition management, and thereby maintain a healthier life. Recently, researchers have been working on smartphone-based diet recording methods and applications that help users accomplish two tasks: record what they eat and how much they eat. Although the former task has made great progress through adopting image recognition technology, it is still a challenge to estimate the volume of foods accurately and conveniently. In this paper, we propose a novel method, named MUSEFood, for food volume estimation. MUSEFood uses the camera to capture photos of the food, but unlike existing volume measurement methods, MUSEFood requires neither training images with volume information nor placing a reference object of known size while taking photos. In addition, considering the impact of different containers on the contour shape of foods, MUSEFood uses a multi-task learning framework to improve the accuracy of food segmentation, and uses a differential model applicable for various containers to further reduce the negative impact of container differences on volume estimation accuracy. Furthermore, MUSEFood uses the microphone and the speaker to accurately measure the vertical distance from the camera to the food in a noisy environment, thus scaling the size of food in the image to its actual size. The experiments on real foods indicate that MUSEFood outperforms state-of-the-art approaches, and highly improves the speed of food volume estimation.

关键词： Image segmentation Containers Shape Sensors Volume measurement Smart phones Solid modeling

来源：评论

学校读者我要写书评

暂无评论

Joint fronthaul multicast and cooperative beamforming for cache-enabled cloud-based small cell networks: An MDS codes-aided approach

arXiv

引用

arXiv 2019年

作者： Wu, Xiongwei Li, Qiang Leung, Victor C.M. Ching, P.C. Department of Electronic Engineering Faculty of Engineering Chinese University of Hong Kong Shatin Hong Kong School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu611731 China College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Department of Electrical and Computer Engineering University of British Columbia VancouverBCV6T 1Z4 Canada

The performance of cloud-based small cell networks (C-SCNs) relies highly on a capacity-limited fronthaul, which degrade quality of service when it is saturated. Coded caching is a promising approach to addressing these challenges, as it provides abundant opportunities for fronthaul multicast and cooperative transmissions. This paper investigates a cache-enabled C-SCNs, in which small-cell base stations (SBSs) are connected to the central processor via fronthaul, and can prefetch popular contents by applying maximum distance separable (MDS) codes. To fully capture the benefits of fronthaul multicast and cooperative transmissions, an MDS codes-aided transmission scheme is first proposed. We formulate the problem to minimize the content delivery latency by jointly optimizing fronthaul bandwidth allocation, SBS clustering, and beamforming. To efficiently solve the resulting nonlinear integer programming problem, we propose a penalty-based design by leveraging variational reformulations of binary constraints. To improve the solution of the penalty-based design, a greedy SBS clustering design is also developed. Furthermore, closed-form characterization of the optimal solution is obtained, through which the benefits of MDS codes can be quantified. Simulation results are given to demonstrate the significant benefits of the proposed MDS codes-aided transmission scheme. Copyright © 2019, The Authors. All rights reserved.

关键词： Multicasting

来源：评论

学校读者我要写书评

暂无评论

Ontology based global and collective motion patterns for event classification in basketball videos

arXiv

引用

arXiv 2019年

作者： Wu, Lifang Yang, Zhou He, Jiaoyu Jian, Meng Xu, Yaowen Xu, Dezhong Chen, Chang Wen College of Information and Communication Engineering Faculty of Information Technology Beijing University of Technology Beijing100124 China Beijing Municipal Key Lab of Computation Intelligence and Intelligent Systems Beijing University of Technology Beijing100124 China School of Science and Engineering Chinese University of Hong Kong Shenzhen518172 China Department of Computer Science and Engineering SUNY Buffalo NY United States

In multi-person videos, especially team sport videos, a semantic event is usually represented as a confrontation between two teams of players, which can be represented as collective motion. In broadcast basketball videos, specific camera motions are used to present specific events. Therefore, a semantic event in broadcast basketball videos is closely related to both the global motion (camera motion) and the collective motion. A semantic event in basketball videos can be generally divided into three stages: pre-event, event occurrence (event-occ), and post-event. By analyzing the influence of different stages of video segments to semantic events discrimination, it is observed that pre-event and event-occ segments are effective for classification, while post-events are effective for event success/failure classification. In this paper, we propose an ontology-based global and collective motion pattern (On_GCMP) algorithm for basketball event classification. First, a two-stage GCMP based event classification scheme is proposed. The GCMP is extracted using optical flow. The two-stage scheme progressively combines a five-class event classification algorithm on event-occs and a two-class event classification algorithm on pre-events. Both algorithms utilize sequential convolutional neural networks (CNNs) and long short-term memory (LSTM) networks to extract the spatial and temporal features of GCMP for event classification. Second, we utilize post-event segments to predict success/failure using deep features of images in the video frames (RGB_DF_VF) based algorithms. Finally the event classification results and success/failure classification results are integrated to obtain the final results. To evaluate the proposed scheme, we collected a new dataset called NCAA+, which is automatically obtained from the NCAA dataset by extending the fixed length of video clips forward and backward of the corresponding semantic events. The experimental results demonstrate that the proposed

关键词： Sports

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：