检索结果-内蒙古大学图书馆

Proceedings of the 22nd International Conference on Neural Information Processing Systems

作者： Guillaume Obozinski Martin J. Wainwright Michael I. Jordan Department of Statistics UC Berkeley Department of Statistics Dept. of Electrical Engineering and Computer Science UC Berkeley Department of Statistics Department of Electrical Engineering and Computer Science UC Berkeley

ISBN: (纸本)9781605609492

We study the behavior of block ℓ1/ℓ2 regulation for multivariate regression, where a K-dimensional response vector is regressed upon a fixed set of p covariates. The problem of support union recovery is to recover the subset of covariates that are active in at least one of the regression problems. Studying this problem under high-dimensional scaling (where the problem parameters as well as sample size n tend to infinity simultaneously), our main result is to show that exact recovery is possible once the order parameter given by θℓ1/ℓ2(n, p, s) : = n/ [2Ѱ(B*) log(p - s)] exceeds a critical threshold. Here n is the sample size, p is the ambient dimension of the regression model, s is the size of the union of supports, and Ѱ(B*) is a sparsity-overlap function that measures a combination of the sparsities and overlaps of the K-regression coefficient vectors that constitute the model. This sparsity-overlap function reveals that block ℓ1/ℓ2 regulation for multivariate regression never harms performance relative to a naive ℓ1-approach, and can yield substantial improvements in sample complexity (up to a factor of K) when the regression vectors are suitably orthogonal relative to the design. We complement our theoretical results with simulations that demonstrate the sharpness of the result, even for relatively small problems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An intrusion detection system for wireless process control systems

An intrusion detection system for wireless process control s...

引用

IEEE Internatonal Conference on Mobile Adhoc and Sensor Systems (MASS)

作者： Tanya Roosta Dennis K. Nilsson Ulf Lindqvist Alfonso Valdes Department of Electrical Engineering and Computer Science UC-Berkeley Department of Computer Science and Engineering Chalmers University of Technology Sweden Computer Science Laboratory SRI International

A recent trend in the process control system (PCS) is to deploy sensor networks in hard-to-reach areas. Using wireless sensors greatly decreases the wiring costs and increases the volume of data gathered for plant monitoring. However, ensuring the security of the deployed sensor network, which is part of the overall security of PCS, is of crucial importance. In this paper, we design a model-based intrusion detection system (IDS) for sensor networks used for PCS. Given that PCS tends to have regular traffic patterns and a well-defined request-response communication, we can design an IDS that models normal behavior of the entities and detects attacks when there is a deviation from this model. Model-based IDS can prove useful in detecting unknown attacks.

关键词： Monitoring SCADA systems Wireless sensor networks Wireless communication Process control Security Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

Union support recovery in high-dimensional multivariate regression

Union support recovery in high-dimensional multivariate regr...

引用

Annual Allerton Conference on Communication, Control, and Computing

作者： Guillaume Obozinski Martin J. Wainwright Michael I. Jordan Department of Statistics and Department of Electrical Engineering and Computer Science UC Berkeley Berkeley

ISBN: (纸本)9781424429257

In the problem of multivariate regression, a K-dimensional response vector is regressed upon a common set of p covariates, with a matrix B~* ∈ R~(p×K) of regression coefficients. We study the behavior of the group Lasso using l_1/l_2 regularization for the union support problem, meaning that the set of s rows for which B~* is non-zero is recovered exactly. Studying this problem under high-dimensional scaling, we show that group Lasso recovers the exact row pattern with high probability over the random design and noise for scalings of (n, p, s) such that the sample complexity parameter given by θ(n,p, s) : = n/[2ψ(B~*) log(p - s)] exceeds a critical threshold. Here n is the sample size, p is the ambient dimension of the regression model, s is the number of non-zero rows, and ψ(B~*) is a sparsity-overlap function that measures a combination of the sparsities and overlaps of the K-regression coefficient vectors that constitute the model. This sparsity-overlap function reveals that, if the design is uncorrelated on the active rows, block l_1/l_2 regularization for multivariate regression never harms performance relative to an ordinary Lasso approach, and can yield substantial improvements in sample complexity (up to a factor of K) when the regression vectors are suitably orthogonal. For more general designs, it is possible for the ordinary Lasso to outperform the group Lasso.

关键词： Multivariate regression Vectors Predictive models Additive noise H infinity control Statistics Size measurement Large-scale systems Statistical learning Collaborative work

来源：评论

学校读者我要写书评

暂无评论

Testbed Implementation of a Secure Flooding Time Synchronization Protocol

Testbed Implementation of a Secure Flooding Time Synchroniza...

引用

IEEE Conference on Wireless Communications and Networking

作者： Tanya Roosta Wei-Chieh Liao Wei-Chung Teng Shankar Sastry Department of Electrical Engineering and Computer Science UC-Automation Berkeley USA Department of Computer Science and Information Engineering National Taiwan University of Science and Technology Taiwan

A fundamental building block in distributed wireless sensor networks is time synchronization. Given resource constrained nature of sensor networks, previous research has focused on developing various energy efficient time synchronization protocols tailored for these networks. However, many of these protocols have not been designed with security in mind. In this paper, we describe FTSP which is one of the major time synchronization protocols for sensor networks. We outline the adverse effects of the time synchronization attacks on some important sensor network applications, and explain the set of possible attacks on FTSP. We then propose a number of countermeasures to mitigate the effect of the security attacks. We implement these attack scenarios on a sensor network testbed and show the extent each attack is successful in desynchronizing the network. Finally, we implement the countermeasures on our sensor network testbed to validate their usefulness in mitigating security attacks. We show that adding a sequence number filter to the original FTSP helps mitigate the effect of attacks on this protocol.

关键词： Testing Floods Protocols Chemical and biological sensors Intelligent sensors Biosensors State estimation Synchronization Information security computer science

来源：评论

学校读者我要写书评

暂无评论

Evaluating a QoS-supportive MAC layer protocol for WLANs

Evaluating a QoS-supportive MAC layer protocol for WLANs

引用

2008 IEEE Military Communications Conference, MILCOM 2008 - Assuring Mission Success

作者： Sarkar, Mahasweta Ray, Moumita Nuggehalli, Pavan Electrical and Computer Engineering San Diego State University San Diego CA 92182 United States Department of Computer Science San Diego State University San Diego CA 92182 United States 2 UC San Diego San Diego CA 92093 United States

ISBN: (纸本)9781424426775

In spite of the tremendous boom in the wireless LAN arena, applications such as streaming video and audio still bring unique and challenging quality of service (QoS) requirements to such networks. To this effect, IEEE 802.11e medium access control (MAC) standard specifies a set of QoS enhancement features which allows traffic classification (High Priority or HP and Low Priority or LP), prioritization and support for various classes of service (preferential and non-preferential service)for various traffic types over a wireless LAN. Thus HP traffic enjoys privileged service hence encounters lesser delay than an application which declares its traffic to be of type LP. However, it is possible for an application to falsely classify its LP traffic as HP traffic in pursuit of preferential service. Such selfish behavior by applications may destroy the QoS capabilities of the network. Currently, the standard does not address this issue. To provide relevant incentives to applications to behave truthfully, a new MAC protocol has been designed which promises fair sharing of resources amongst both HP and LP traffic such that the "utility" of both traffic classes is maximized. This paper presents an evaluation of this new MAC protocol through simulations and analysis results. Specifically, this paper analyses the incentive mechanism proposed in this MAC and studies its effect on system performance. It also compares this MAC scheme to a scheme similar to that suggested in IEEE 802.11e. Finally, a QualNet simulation is provided which demonstrates the feasibility of incorporating our scheme in the current IEEE 802.11e standard and the efficiency of our scheme over the same in the presence of selfish users. ©2008 IEEE.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Structural perfection of InGaN layers and its relation to photoluminescence

引用

physica status solidi c 2009年第12期6卷

作者： Z. Liliental-Weber K. M. Yu M. Hawkridge S. Bedair A.E. Berman A. Emara D. R. Khanal J. Wu J. Domagala J. Bak-Misiuk Materials Science Division Lawrence Berkeley National Laboratory m/s 62R203-8255 Berkeley CA 94720 USA Phone: +01 510 486 6656 Fax: +01 510 486 5530 Electrical and Computer Engineering Department North Carolina State University Raleigh NC 27695 USA Department of Materials Science and Engineering UC Berkeley CA 94720 USA Institute of Physics Polish Academy of Sciences Al. Lotnikow 32/46 02-668 Warsaw Poland

The relation between structural perfection and optical properties of InGaN with 10% In are discussed. Transmission Electron Microscopy, X-ray diffraction and Rutherford backscattering spectrometry measurements show that only strained layers with a thickness not exceeding 100 nm are defect free and In concentration is lower than the nominal value. Extension of layer thickness leads to layer sequestration into sublayers with different In contents and the formation of planar defects as a result of layer relaxation. In concentration in such sublayers reach and in some cases exceed the nominal concentration. A single band edge photoluminescence peak is observed only for the thinnest layer. Samples with larger film thickness showed multiple PL peaks corresponding to layers with different In content. Much higher In content would be required to explain the presence of some PL peaks, suggesting that some PL peaks originate from the defective areas of the film. This was confirmed by cathodoluminescence studies performed on the same samples used earlier for TEM studies. (© 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

关键词： 61.05.cp 61.72.Nn 61.72.Qq 68.35.Dv 68.37.Lp 78.55.Cr

来源：评论

学校读者我要写书评

暂无评论

ICMLA 2008 Invited Speakers

ICMLA 2008 Invited Speakers

引用

International Conference on Machine Learning and Applications (ICMLA)

作者： Dan Roth Jude Shavlik Michael I. Jordan Bin Yu Andrew Moore Philip Bourne Department of Computer Science University of Illinois at Urbana Champaign University of Wisconsin Madison Department of Electrical Engineering and Computer Science University of California Berkeley Department of Statistics Department of EECS UC Berkeley Google Pittsburgh University of California San Diego

Provides an abstract for each of the invited presentations and a brief professional biography of each presenter. The complete presentations were not made available for publication as part of the conference proceedings.

关键词： Natural languages Machine learning computer science Data mining Statistical analysis Data structures Brain modeling Space technology Natural language processing Machine learning algorithms

来源：评论

学校读者我要写书评

暂无评论

The WSU model for engineering mathematics education: A multiyear assessment and expansion to collaborating institutions

The WSU model for engineering mathematics education: A multi...

引用

2008 ASEE Annual Conference and Exposition

作者： Klingbeil, Nathan Rattan, Kuldip Raymer, Michael Reynolds, David Mercer, Richard Kukreti, Anant Randolph, Brian Wright State University Department of Mechanical Engineering Department of Teaching Wright State University WSU's National Model for Engineering Mathematics Education Department of Electrical Engineering Wright State University Department of Computer Science and Engineering Wright State University Department of Biomedical Industrial and Human Factors Engineering Wright State University Department of Mathematics and Statistics Wright State University University of Cincinnati Engineering Education Research UC Adoption WSU's National Model for Engineering Mathematics Education University of Toledo Undergraduate Studies Department of Civil Engineering University of Toledo UT Adoption WSU's National Model for Engineering Mathematics Education

The inability of incoming students to advance past the traditional first-year calculus sequence is a primary cause of attrition in engineering programs across the country. As a result, this paper will describe an NSF funded initiative at Wright State University to redefine the way engineering mathematics is taught, with the goal of increasing student retention, motivation and success in engineering. The WSU approach begins with the development of a novel first-year engineering mathematics course, EGR 101 "Introductory Mathematics for engineering Applications." Taught by engineering faculty, the course includes lecture, laboratory and recitation components. Using an application-oriented, hands-on approach, the course addresses only the salient math topics actually used in core engineering courses. These include the traditional physics, engineering mechanics, electric circuits and computer programming sequences. The EGR 101 course replaces traditional math prerequisite requirements for the above core courses, so that students can advance in the curriculum without having completed a traditional first-year calculus sequence. The WSU model concludes with a revised engineering math sequence, taught by the math department later in the curriculum, in concert with College and ABET requirements. The result has shifted the traditional emphasis on math prerequisite requirements to an emphasis on engineering motivation for math, with a "just-in-time" structuring of the required math sequence. This paper includes significant updates since the approach was last reported, including a multiyear assessment at Wright State University and expansion of the program to collaborating institutions. © American Society for engineering Education, 2008.

关键词： Curricula

来源：评论

学校读者我要写书评

暂无评论

Subspace-based methods for image registration and super-resolution

Subspace-based methods for image registration and super-reso...

引用

IEEE International Conference on Image Processing

作者： Patrick Vandewalle Loic Baboulaz Pier Luigi Dragotti Martin Vetterli Philips Research Eindhoven The Netherlands Electrical and Electronic Engineering Department Imperial College London UK Department of Electrical Engineering and Computer Science UC Berkeley USA Ecole Polytechnique Fédérale de Lausanne School of Computer and Communication Sciences Switzerland

Super-resolution algorithms combine multiple low resolution images into a single high resolution image. They have received a lot of attention recently in various application domains such as HDTV, satellite imaging, and video surveillance. These techniques take advantage of the aliasing present in the input images to reconstruct high frequency information of the resulting image. One of the major challenges in such algorithms is a good alignment of the input images: subpixel precision is required to enable accurate reconstruction. In this paper, we give an overview of some subspace techniques that address this problem. We first formulate super-resolution in a multichannel sampling framework with unknown offsets. Then, we present three registration methods: one approach using ideas from variable projections, one using a Fourier description of the aliased signals, and one using a spline description of the sampling kernel. The performance of the algorithms is evaluated in numerical simulations.

关键词： Image registration Image resolution Signal resolution Image reconstruction HDTV Satellites High-resolution imaging Video surveillance Frequency Image sampling

来源：评论

学校读者我要写书评

暂无评论

Closing the loop on test creation: A question assessment mechanism for instructors

Closing the loop on test creation: A question assessment mec...

引用

37th SIGCSE Technical Symposium on computer science Education

作者： Winters, Titus Payne, Tom Computer Science and Engineering Department UC Riverside Riverside CA 92521

ISBN: (纸本)9781595932594

New accreditation requirements focus on education as a "continuous improvement process." The most important part of such a process is that information gets fed back into the system to improve the quality of the output. This requirement is often interpreted to imply a course-level feedback loop that iterates on offerings of courses or entire academic years. This paper provides a smaller and more immediate feedback loop. This technique gives instructors feedback onthe quality of each question on a test or quiz, as well as a numeric score for the difficult of the question. A simple tool implementing this procedure can be used to help train instructors on which questions are difficult, as well as what types of questions are correlated with ability, and how to design a meaningful instrument of assessment. Performing this analysis at the end of a course offering could help demonstrate continuous improvement to accreditation committees. Performing this analysis immediately after the administration of a test or quiz can point out topics that the class as a whole have failed to understand, thus giving instructors more insight into student knowledge. Copyright 2006 ACM.

关键词： Accreditation

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：