检索结果-内蒙古大学图书馆

25th ACM International Conference on Multimedia, MM 2017

作者： Zhou, Yiyi Ji, Rongrong Su, Jinsong Wu, Yongjian Wu, Yunsheng Fujian Key Laboratory of Sensing and Computing for Smart City Xiamen University 361005 China School of Information Science and Engineering Xiamen University 361005 China School of Software Xiamen University 361005 China Tencent YouTu Lab. Tencent Technology Shanghai China

ISBN: (纸本)9781450349062

Most of existing works in visual question answering (VQA) are dedicated to improving the performance of answer predictions, while leaving the explanation of answering unexploited. We argue that, exploiting the explanations of question answering not only makes VQA explainable, but also quantitatively improves the prediction performance. In this paper, we propose a novel network architecture, termed Neural Pivot Network (NPN), towards simultaneous VQA and generating explanations in a multi-task learning architecture. NPN is trained by using both image-caption and image-question-answer pairs. In principle, CNN-based deep visual features are extracted and sent to both the VQA channel and the captioning module, the latter of which serves as a pivot to bridge the source image module to the target QA predictor. Such an innovative design enables us to introduce large-scale image-captioning training sets, e.g., MS-COCO Caption and Visual Genome Caption, together with cutting-edge image captioning models to benefit VQA learning. Quantitatively, the proposed NPN performs significantly better than alternatives and state-of-the-art schemes trained on VQA datasets only. Besides, by investigating the by-product of experiments, in-depth digests can be provided along with the answers. © 2017 ACM.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Perturbation techniques for convergence analysis of proximal gradient method and other first-order algorithms via variational analysis

arXiv

引用

arXiv 2018年

作者： Wang, Xiangfeng Ye, Jane J. Yuan, Xiaoming Zeng, Shangzhi Zhang, Jin Shanghai Key Lab for Trustworthy Computing School of Computer Science and Software Engineering East China Normal University China Department of Mathematics and Statistics University of Victoria Canada Department of Mathematics University of Hong Kong Hong Kong Hong Kong Department of Mathematics University of Hong Kong Hong Kong Hong Kong Department of Mathematics Hong Kong Baptist University Hong Kong HKBU Institute of Research and Continuing Education Shenzhen China

We develop new perturbation techniques for conducting convergence analysis of various first-order algorithms for a class of nonsmooth optimization problems. We consider the iteration scheme of an algorithm to construct a perturbed stationary point set-valued map, and define the perturbing parameter by the difference of two consecutive iterates. Then, we show that the calmness condition of the induced set-valued map, together with a local version of the proper separation of stationary value condition, is a sufficient condition to ensure the linear convergence of the algorithm. The equivalence of the calmness condition to the one for the canonically perturbed stationary point set-valued map is proved, and this equivalence allows us to derive some sufficient conditions for calmness by using some recent developments in variational analysis. These sufficient conditions are different from existing results (especially, those error-bound-based ones) in that they can be easily verified for many concrete application models. Our analysis is focused on the fundamental proximal gradient (PG) method, and it enables us to show that any accumulation of the sequence generated by the PG method must be a stationary point in terms of the proximal subdifferential, instead of the limiting subdifferential. This result finds the surprising fact that the solution quality found by the PG method is in general superior. Our analysis also leads to some improvement for the linear convergence results of the PG method in the convex case. The new perturbation technique can be conveniently used to derive linear rate convergence of a number of other first-order methods including the well-known alternating direction method of multipliers and primal-dual hybrid gradient method, under mild assumptions. Copyright © 2018, The Authors. All rights reserved.

关键词： Gradient methods

来源：评论

学校读者我要写书评

暂无评论

On the Convergence Rate of a Class of Proximal-Based Decomposition Methods for Monotone Variational Inequalities

引用

Journal of the Operations Research Society of China 2015年第3期3卷 347-362页

作者： Xiang-Feng Wang Shanghai Key Lab for Trustworthy Computing Software Engineering InstituteEast China Normal UniversityShanghai 200062China

A unified efficient algorithm framework of proximal-based decomposition methods has been proposed for monotone variational inequalities in 2012,while only global convergence is proved at the same *** this paper,we give a unified proof on theO(1/t)iteration complexity,together with the linear convergence rate for this kind of proximal-based decomposition *** theε-optimal iteration complexity result defined by variational inequality,the non-ergodic relative error of adjacent iteration points is also proved to decrease in the same ***,the linear convergence rate of this algorithm framework can be constructed based on some special variational inequality properties,without necessary strong monotone conditions.

关键词： Variational inequality Proximal point algorithm Iteration complexity Relative error Convergence rate Error bound

来源：评论

学校读者我要写书评

暂无评论

More Than Address: Pre-Identify Your Income with the Open Data

More Than Address: Pre-Identify Your Income with the Open Da...

引用

International Conference on Cloud computing and Big Data, CCBD 2015

作者： Li, Ruoyu Xiong, Huanyu Zhao, Hui Software Engineering Institute East China Normal University Shanghai China Shanghai Key Lab. of Trustworthy Computing East China Normal University China

ISBN: (纸本)9781467383509

Modern e-commerce websites collect immense amounts of transaction data, and almost each of them is associated with at least one delivery address. To promote sales, e-commerce operators are eager to identify the customers consumption levels. However, the information resource for target customers is quite limited due to privacy protection. In this paper, we propose an integrated framework together with an empirical method to pre-identify customers purchasing ability. To implement that, we need to recognize the implicit information within the delivery addresses of customers. In practice, both transaction data from the e-commerce providers and open real estate data associated with the delivery addresses are made rational utilization of. The experimental results show that this method is a feasible solution for income segment prediction, which can be extensively applied to marketing, advertising and product promotion, etc. © 2015 IEEE.

关键词： Sales

来源：评论

学校读者我要写书评

暂无评论

A validated method of solving initial value problem for ODEs

A validated method of solving initial value problem for ODEs

引用

Proceedings of the Asia-Pacific Conference on Computer Science and Applications, CSAC 2014

作者： Tang, Min Wu, Min Deng, Guoqiang Shanghai Key Lab of Trustworthy Computing Software Engineering Institute East China Normal University Shanghai China School of Mathematics & Computing Science Guilin University of Electrical Technology Guilin China

ISBN: (纸本)9781138028111

In this paper, we present an interval method to test existence and uniqueness of solutions to the initial value problem of ODEs, and an iterative computational test is proposed to compute validated solutions of ODEs. Compared with classical methods, the interval method not only returns an approximate solution but also an interval that is guaranteed to contain exact solutions of the ODEs. The experiments show the effectiveness of the validated method. © 2015 Taylor & Francis Group, London.

关键词： Initial value problems

来源：评论

学校读者我要写书评

暂无评论

Joint Power Splitting and Secure Beamforming Design in the Wireless-powered Untrusted Relay Networks

Joint Power Splitting and Secure Beamforming Design in the W...

引用

IEEE Global Communications Conference

作者： Mingxiong Zhao Suili Feng Xiangfeng Wang Meng Zhang Yuan Liu Hao Fu School of Electronic and Information Engineering South China University of Technology Shanghai Key Lab. of Trustworthy Computing Software Engineering Institute East China Normal Univ.

ISBN: (纸本)9781479959532

In this work, we maximize the secrecy rate of the wireless-powered untrusted amplify-and-forward relay networks by jointly designing power splitting (PS) ratio and relay beamforming with the proposed global optimal algorithm (GOA) and local optimal algorithm (LOA). To guarantee secure communication, the destination-based artificial noise is sent to degrade the reception of the untrusted relay, and it also becomes a new source of energy powering relay to forward the information with power splitting (PS) technique. Simulation result shows that LOA can achieve satisfactory secrecy rate performance compared with that of GOA, but with less computation time. It also manifests that both proposed algorithms outperform the benchmark method.

关键词： Relay networks power splitting TRIM47 gene Loa computational time Beamforming reception relay Privacy

来源：评论

学校读者我要写书评

暂无评论

More Than Address: Pre-Identify Your Income with the Open Data

More Than Address: Pre-Identify Your Income with the Open Da...

引用

International Conference on Cloud computing and Big Data (CCBD)

作者： Ruoyu Li Huanyu Xiong Hui Zhao Software Engineering Institute East China Normal University Shanghai China Shanghai Key Lab of Trustworthy Computing East China Normal University China

关键词： Companies Medical services Databases Standards Privacy Data mining History

来源：评论

学校读者我要写书评

暂无评论

On deploying relays for connected indoor sensor networks

引用

Journal of Communications and Networks 2014年第3期16卷 335-343页

作者： Zhu, Yanmin Xue, Cuiyao Cai, Haibin Yu, Jiadi Ni, Lei Li, Minglu Li, Bo Shanghai Key Lab. of Scalable Computing and Systems Department of Computer Science and Engineering Shanghai Jiao Tong University China Department of Computer Science and Engineering Shanghai Jiao Tong University China Software Engineering Institute East China Normal University China Shanghai Jiao Tong University Hong Kong University of Science and Technology Hong Kong

This paper considers the crucial problem of deploying wireless relays for achieving a connected wireless sensor network in indoor environments, an important aspect related to the management of the sensor network. Several algorithms have been proposed for ensuring full sensing coverage and network connectivity. These algorithms are not applicable to indoor environments because of the complexity of indoor environments, in which a radio signal can be dramatically degraded by obstacles such as walls. We first prove theoretically that the indoor relay placement problem is NP-hard. We then predict the radio coverage of a given relay deployment in indoor environments. We consider two practical scenarios;wire-connected relays and radio-connected relays. For the network with wire-connected relays, we propose an efficient greedy algorithm to compute the deployment locations of relays for achieving the required coverage percentage. This algorithm is proved to provide a Hn factor approximation to the theoretical optimum, where Hn = 1 +1/2+⋯+1/n= ln(n) +1, and n is the number of all grid points. In the network with radio-connected relays, relays have to be connected in an ad hoc mode. We then propose an algorithm based on the previous algorithm for ensuring the connectivity of relays. Experimental results demonstrate that the proposed algorithms achieve better performance than baseline algorithms. © 2014 KICS.

关键词： Approximation algorithms

来源：评论

学校读者我要写书评

暂无评论

Coalgebras for fuzzy transition systems

Coalgebras for fuzzy transition systems

引用

作者： Wu, Hengyang Chen, Yixiang Information Engineer College Hangzhou Dianzi University Hangzhou 310018 China MoE Engineering Research Center for Software/Hardware Co-design Technology and Application Shanghai Key Lab for Trustworthy Computing East China Normal University Shanghai 200062 China

This paper studies a coalgebraic theory of fuzzy transition systems. Main conclusions include: the functor FA for deterministic fuzzy transition systems and the functor (Pa°F)A for nondeterministic fuzzy transition systems preserve weak pullbacks, and the functor FA has a final coalgebra under some restricted conditions. Moreover, we show how to get a concrete (fuzzy) bisimulation from a coalgebraic bisimulation. Crown Copyright © 2014 Published by Elsevier B.V. All rights reserved.

关键词： Algebra

来源：评论

学校读者我要写书评

暂无评论

Comprehensive economic strength assessment based on decision tree algorithm

引用

Computer Modelling and New Technologies 2014年第12期18卷 403-407页

作者： Hou, Yuan Wu, Min Shanghai Key Lab of Trustworthy Computing Software Engineering Institute East China Normal University Shanghai China

In this paper, we apply some computer software techniques in socio-economics to study an economic decision-making problem, the comprehensive economic strength assessment problem. To achieve delicacy socio-economical decision, we first conduct an asses-sment of various influencing factors involved in the decision problem based on analytic hierarchy process, and then apply two deci-sion-tree algorithms to establish the appropriate models with corresponding classification rules. We also use Matlab.to fit the data for the purpose of prediction and validation. The experiments show the effectivity and accuracy of decision-tree algorithms in supplying valuable suggestions for economic decision-making problems.

关键词： Data mining

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：