检索结果-内蒙古大学图书馆

The Journal of Machine Learning Research 2013年第1期14卷

作者： Kevin Murphy Bernhard Schölkopf Yu-Feng Li Ivor W. Tsang James T. Kwok Zhi-Hua Zhou Google MPI for Intelligent Systems National Key Laboratory for Novel Software Technology Nanjing University Nanjing China School of Computer Engineering Nanyang Technological University Singapore Department of Computer Science and Engineering Hong Kong University of Science & Technology Hong Kong

In this paper, we study the problem of learning from weakly labeled data, where labels of the training examples are incomplete. This includes, for example, (i) semi-supervised learning where labels are partially known; (ii) multi-instance learning where labels are implicitly known; and (iii) clustering where labels are completely unknown. Unlike supervised learning, learning with weak labels involves a difficult Mixed-Integer Programming (MIP) problem. Therefore, it can suffer from poor scalability and may also get stuck in local minimum. In this paper, we focus on SVMs and propose the WELLSVM via a novel label generation strategy. This leads to a convex relaxation of the original MIP, which is at least as tight as existing convex Semi-Definite Programming (SDP) relaxations. Moreover, the WELLSVM can be solved via a sequence of SVM subproblems that are much more scalable than previous convex SDP relaxations. Experiments on three weakly labeled learning tasks, namely, (i) semi-supervised learning; (ii) multi-instance learning for locating regions of interest in content-based information retrieval; and (iii) clustering, clearly demonstrate improved performance, and WELLSVM is also readily applicable on large data sets.

关键词： clustering convex relaxation cutting plane multi-instance learning semi-supervised learning weakly labeled data

来源：评论

学校读者我要写书评

暂无评论

Comparing six modeling approaches

引用

Workshops and Symposia on Models in software engineering, MODELS 2011

作者： Mussbacher, Gunter Al Abed, Wisam Alam, Omar Ali, Shaukat Beugnard, Antoine Bonnet, Valentin Bræk, Rolv Capozucca, Alfredo Cheng, Betty H.C. Fatima, Urooj France, Robert Georg, Geri Guelfi, Nicolas Istoan, Paul Jézéquel, Jean-Marc Kienzle, Jörg Klein, Jacques Lézoray, Jean-Baptiste Malakuti, Somayeh Moreira, Ana Phung-Khac, An Troup, Lucy Department of Systems and Computer Engineering Carleton University Canada Software Engineering Laboratory McGill University Montreal Canada Simula Research Laboratory Norway Département Informatique TELECOM Bretagne France Department of Telematics Norwegian University of Science and Technology Norway LASSY Research Team University of Luxembourg Luxembourg Department of Computer Science and Engineering Michigan State University United States Computer Science Department Colorado State University Fort Collins United States ISC Department CRP Gabriel Lippmann Luxembourg INRIA Centre Rennes - Bretagne Atlantique Rennes France Laboratory for Advanced Software Systems University of Luxembourg Luxembourg Software Engineering Group University of Twente Netherlands Universidade Nova de Lisboa Lisbon Portugal Department of Psychology Colorado State University Fort Collins United States

ISBN: (纸本)9783642296444

While there are many aspect-oriented modeling (AOM) approaches, from requirements to low-level design, it is still difficult to compare them and know under which conditions different approaches are most applicable. This comparison, however, is crucially important to unify existing AOM and more traditional object-oriented modeling (OOM) approaches and to generalize individual approaches into a comprehensive end-to-end method. Such a method does not yet exist. This paper reports on work done at the inaugural Comparing Modeling Approaches (CMA) workshop towards the goal of identifying potential comprehensive methodologies: (i) a common, focused case study for six modeling approaches, (ii) a set of criteria applied to each of the six approaches, and (iii) the assessment results. © 2012 Springer-Verlag Berlin Heidelberg.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Reliable Provisioning of Spot Instances for Compute-intensive Applications

Reliable Provisioning of Spot Instances for Compute-intensiv...

引用

International Conference on Advanced Information Networking and Applications (AINA)

作者： William Voorsluys Rajkumar Buyya Cloud Computing and Distributed Systems (CLOUDS) Laboratory Department of Computer Science and Software Engineering University of Melbourne Australia

Cloud computing providers are now offering their unused resources for leasing in the spot market, which has been considered the first step towards a full-fledged market economy for computational resources. Spot instances are virtual machines (VMs) available at lower prices than their standard on-demand counterparts. These VMs will run for as long as the current price is lower than the maximum bid price users are willing to pay per hour. Spot instances have been increasingly used for executing compute-intensive applications. In spite of an apparent economical advantage, due to an intermittent nature of biddable resources, application execution times may be prolonged or they may not finish at all. This paper proposes a resource allocation strategy that addresses the problem of running compute-intensive jobs on a pool of intermittent virtual machines, while also aiming to run applications in a fast and economical way. To mitigate potential unavailability periods, a multifaceted fault-aware resource provisioning policy is proposed. Our solution employs price and runtime estimation mechanisms, as well as three fault-tolerance techniques, namely check pointing, task duplication and migration. We evaluate our strategies using trace-driven simulations, which take as input real price variation traces, as well as an application trace from the Parallel Workload Archive. Our results demonstrate the effectiveness of executing applications on spot instances, respecting QoS constraints, despite occasional failures.

关键词： Fault tolerance Fault tolerant systems Checkpointing Runtime Virtual machining Computational modeling

来源：评论

学校读者我要写书评

暂无评论

Preemption-aware Admission Control in a Virtualized Grid Federation

Preemption-aware Admission Control in a Virtualized Grid Fed...

引用

International Conference on Advanced Information Networking and Applications (AINA)

作者： Mohsen Amini Salehi Bahman Javadi Rajkumar Buyya Cloud Computing and Distributed Systems (CLOUDS) Laboratory Department of Computer Science and Software Engineering University of Melbourne Australia

Many applications in federated Grids have quality-of-service (QoS) constraints such as deadline. Admission control mechanisms assure QoS constraints of the applications by limiting the number of user requests accepted by a resource provider. However, in order to maximize their profit, resource owners are interested in accepting as many requests as possible. In these circumstances, the question that arises is: what is the effective number of requests that can be accepted by a resource provider in a way that the number of accepted external requests is maximized and, at the same time, QoS violations are minimized. In this paper, we answer this question in the context of a virtualized federated Grid environment, where each Grid serves requests from external users along with its local users and requests of local users have preemptive priority over external requests. We apply analytical queuing model to address this question. Additionally, we derive a preemption-aware admission control policy based on the proposed model. Simulation results under realistic working conditions indicate that the proposed policy improves the number of completed external requests (up to 25%). In terms of QoS violations, the 95% confidence interval of the average difference with other policies is between (14.79%, 18.56%).

关键词： Admission control Mathematical model Equations Analytical models Time factors Queueing analysis

来源：评论

学校读者我要写书评

暂无评论

Enhancing the system development process performance: A value-based approach

Enhancing the system development process performance: A valu...

引用

22nd Annual International Symposium of the International Council on systems engineering, INCOSE 2012 and the 8th Biennial European systems engineering Conference 2012, EuSEC 2012

作者： Mastrofmi, Manuel Cantone, Giovanni Seaman, Carolyn Shull, Forrest Diep, Madeline Falessi, Davide University of Rome Tor Vergata Department of Computer Science and Systems and Production Via del Politecnico 1 00133 Rome Italy Fraunhofer Center for Experimental Software Engineering University of Maryland Baltimore County Department of Information Systems ITE. Building. 1000 Hilltop Circle Baltimore MD 21250 United States Fraunhofer Center for Experimental Software Engineering 5825 University Research Court Suite 1300 College Park MD 20740-3823 United States Simula Research Laboratory Oslo Norway University of Rome Tor Vergata Department of Computer Science Systems and Production Via del Politecnico 1 00133 Rome Italy

ISBN: (纸本)9781622769162

When planning or controlling the system development process, a project leader needs to make decisions which take into account a number of aspects, including: availability of assets and competences, previously enacted processes in the organization, certifications the system is required to obtain, standards to comply with, interactions among process activities, contextual factors and constraints, and allocated budget and schedule. In this paper we propose a value-based approach for supporting decision making. The aim is to provide supportive information for decisions related to the system verification process. This would in turn enhance the performance of system development process by supporting the decision making process for complex systems. We report both academic and industrial empirical evaluations, which demonstrate the feasibility and effectiveness of our proposal, and thus prompt us to refine and extend our approach to sub-processes other than verification. ©2012 by the authors.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Kernel methods based approaches to image classification and retrieval

Kernel methods based approaches to image classification and ...

引用

International Conference on Advanced Computing, ICAC

作者： C. Chandra Sekhar Sasi Kumar Madhan Subhas Raj Kumar Buyya Department of Computer Science and Engineering Indian Institute of Technology Madras CDAC Pune Chronus Corporation Computer Science and Software Engineering and Director of the Cloud Computing and Distributed Systems (CLOUDS) Laboratory at the University of Melbourne Australia

Support vector machine (SVM) is a pattern classification model suitable for classification and annotation of images using non-vectorial type representations of images. Varying length patterns extracted from image data correspond to sets of local feature vectors. Kernels designed for varying length patterns are called as dynamic kernels. The talk presents the issues in designing the dynamic kernel based SVMS for classification and annotation of images. Different methods for designing the dynamic kernels are presented. An intermediate matching kernel (IMK) for a pair of varying length patterns is constructed by matching the pairs of local feature vectors selected using a set of virtual feature vectors. For patterns corresponding to sets of feature vectors, a Gaussian mixture model (GMM) is used as the set of virtual feature vectors. The GMM-based IMK is considered for image processing tasks such as image classification, matching and annotation in content-based image retrieval. The talk presents results of experimental studies on image classification, annotation and retrieval of images using the kernel methods.

关键词： Cloud computing Kernel Technological innovation Support vector machine classification Learning systems Hidden Markov models Computational modeling

来源：评论

学校读者我要写书评

暂无评论

Faculty development in the EU ERAMIS project

Faculty development in the EU ERAMIS project

引用

2012 IEEE Global engineering Education Conference, EDUCON 2012

作者： Merceron, Agathe Adam, Jean-Michel Lujan-Mora, Sergio Milosz, Marek Toppinen, Arto Media Informatics Department Beuth University of Applied Sciences Berlin Germany Laboratory of Informatics of Grenoble Pierre Mendes France University Grenoble France Department of Software and Computing Systems University of Alicante Alicante Spain Institute of Computer Science Lublin University of Technology Lublin Poland School of Engineering and Technology Savonia University of Applied Sciences Savonia Finland

The aim of the ERAMIS project is to create a Master degree computer as a Second Competence in 9 beneficiary universities of Kazakhstan, Kyrgyzstan and Russia. This contribution presents how faculty development is orga... 详细信息

关键词： Teaching

来源：评论

学校读者我要写书评

暂无评论

High Availability through Output Continuity

High Availability through Output Continuity

引用

IEEE International Conference on Cluster Computing Workshops and Posters (CLUSTER WORKSHOPS)

作者： Wei Ye Yaozu Dong Ruhui Ma Alei Liang Haibing Guan Shanghai Key Laboratory of Scalable Computing and Systems Department of Computer Science and Engineering Shanghai Jiaotong University Shanghai China Intel China Software Center Shanghai China

Virtual machine (VM) based state machine approaches, i.e. VM replication, provide high availability without source code modifications, unfortunately, existing VM replication approaches suffer from excessive replication overheads. On the other hand, machine state replication is an overly strong (and therefore less efficient) requirement for high availability in the networked client-server system. In this paper, we propose a generic and highly efficient high availability solution through output continuity. Output continuity considers a backup server as a valid replica as long as the responses generated so far are the same between the backup and primary servers, the server state is propagated from the primary server to the backup server if and only if the previous outputs from the backup and primary servers are different.

关键词： Servers Availability Mathematical model Semantics Equations Hardware

来源：评论

学校读者我要写书评

暂无评论

An Embedded Co-AdaBoost and Its Application in Classification of software Document Relation

An Embedded Co-AdaBoost and Its Application in Classificatio...

引用

International Conference on Semantics, Knowledge and Grid (SKG)

作者： Jin Liu Juan Li Yuan Xie Jeff Lei Qiping Hu State Key Laboratory of Software Engineering Computer School Wuhan University China State Key Laboratory of Intelligent Control and Management of Complex Systems Institute of Automation CAS China Department of Computer Science and Engineering University of Texas Arlington USA International School of Software Wuhan University China

To enhance classification performance by making use of easily available unlabelled data to overcome the scarcity of labelled data, this paper proposes an Embedded Co-Adaboost algorithm that integrates multi-view learning into the Adaboost learning framework and at the same time leverages the advantages of Co-training algorithm for performance enhancement. Experimental results demonstrate the effectiveness of the proposed algorithm in terms of the convergence rate, the accuracy, and the steady performance as compared to the original AdaBoost algorithm, without relying on redundant and sufficient feature sets. As a algorithm application in software engineering, the Embedded Co-AdaBoost has been applied to the classification of software document relations to improve the quality of the architecture design documents and the reusability of design knowledge.

关键词： Classification algorithms Training software Accuracy Partitioning algorithms software algorithms Convergence

来源：评论

学校读者我要写书评

暂无评论

Bias analyses of spontaneous facial expression database

Bias analyses of spontaneous facial expression database

引用

International Conference on Pattern Recognition

作者： Zhaoyu Wang Shangfei Wang Yachen Zhu Qiang Ji Key Laboratory of Computing and Communicating Software of Anhui Province School of Computer Science and Technology University of Science and Technology of China Hefei Anhui China Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute Troy NY USA

ISBN: (纸本)9781467322164

In this paper, cross-corpora evaluations are used to analyze the bias of spontaneous facial expression databases. Local binary pattern, Gabor, eigenface and fisherface features are extracted and applied to the four spontaneous expression databases: USTC-NVIE, VAM, Belfast Naturalistic and SEMAINE to recognize arousal (high/low) and valance (positive/negative) respectively. Experimental results indicate that there exists bias among different spontaneous expression databases. The emotion-induction methods, the variety of subjects and the quantity of raters may have caused such a bias.

关键词： Databases Feature extraction Face recognition Videos Principal component analysis Training

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：