检索结果-内蒙古大学图书馆

International Joint Conference on Neural Networks (IJCNN)

作者： Gao, Junbin Charles Sturt Univ Sch Comp & Math Bathurst NSW 2795 Australia

ISBN: (纸本)9781424496365

In this paper we propose a nonparametric extension to the sparse kernel machine using a beta process prior. The extended beta process sparse kernel machine (BPSKM) allows for a sparse model to be constructed from a set of training data. The recent research on beta process reveals elegant property of Bayesian conjugate prior which is utilized to derive a variational Bayes inference algorithm. The performance of the proposed algorithm has been investigated on both synthetic and real-life data sets.

关键词： Bayes methods Bayesian conjugate prior Bayesian methods Data models Kernel Machine learning Sparse matrices Training Training data inference mechanisms learning (artificial intelligence) multiprogramming multitask beta process sparse kernel machine nonparametric extension real-life data set sparse matrices variational Bayes inference algorithm

来源：评论

学校读者我要写书评

暂无评论

Wavelet As A Diagnostic Tool For Fault Classification And Identification In Underground Power Cable

Wavelet As A Diagnostic Tool For Fault Classification And Id...

引用

Conference on Electrical Insulation and Dielectric Phenomena (CEIDP)

作者： Pandey, Abhishek Younan, Nicolas Mississippi State Univ Dept Elect & Comp Engn Mississippi State MS 39762 USA

ISBN: (纸本)9781457709869

In this paper, a diagnostic technique based on the wavelet transform (WT) for identifying and classifying different fault types is developed. In general, the wavelet transform provides multiresolution type analysis with variable window size;thus better suited for such analysis. Variable window sizes provide good resolution in the time and frequency domains;thus helping not only in identifying the fault, but also in localizing it. Three types of cables are used in this study;a normal cable, a shorted cable, and a cable with holes. Wavelet transformation is then applied to the impedance in each case, with various windowing to eliminate possible interference. The identification is then carried out by examining the resulting impedance magnitude. This study reveals that the wavelet transform is a better tool for fault identification and classification.

关键词： cable Fault class wavelet transforms wavelets diagnostic tool Underground cables failure recognition multiprogramming Window Size impedance underground

来源：评论

学校读者我要写书评

暂无评论

Exploring Energy-Efficient DRAM Array Organizations

Exploring Energy-Efficient DRAM Array Organizations

引用

54th IEEE International Midwest Symposium on Circuits and Systems (MWSCAS)

作者： O, Seongil Choo, Sungwoo Ahn, Jung Ho Seoul Natl Univ Grad Sch Convergence Sci & Technol Seoul South Korea

ISBN: (纸本)9781612848570

DRAM is widely used as main-memory storage in contemporary computer systems. As VLSI process technology advances, more transistors can be integrated in a single die leading to higher storage capacity and communication throughput per DRAM chip. New DRAM standards are created in order to keep up with these trends, and many factors such as performance, energy efficiency, reliability, and fabrication/testing cost are considered when a new DRAM architecture is designed. However, there are few studies on DRAM array organizations that consider both performance and energy efficiency of entire computer systems using the organizations. In this paper, we explore the design space of contemporary DRAM array organizations by varying the number of pages that can be concurrently accessed and the size of the pages. We compare various main-memory DRAM array organizations using multithreaded and multiprogrammed workloads on a chip-multiprocessor system with die-stacked DRAM memory in search of energy-efficient array configurations.

关键词： Arrays DRAM architecture DRAM chips DRAM chips Energy efficiency Organizations Timing VLSI VLSI process technology Wires chip-multiprocessor system communication throughput contemporary computer system design space die-stacked DRAM memory energy efficiency energy-efficient DRAM array organization energy-efficient array configuration fabrication cost main-memory storage microprocessor chips multi-threading multiprogrammed workload multiprogramming multithreaded workload reliability storage capacity testing cost transistor

来源：评论

学校读者我要写书评

暂无评论

PACMan: Prefetch-aware cache management for high performance caching 44

PACMan: Prefetch-aware cache management for high performance...

引用

44th Annual IEEE/ACM Symposium on Microarchitecture, MICRO 44

作者： Wu, Carole-Jean Jaleel, Aamer Martonosi, Margaret Steely Jr., Simon C. Emer, Joel Princeton University Princeton NJ United States Intel Corporation VSSAD Hudson MA United States Massachusetts Institute of Technology Cambridge MA United States

ISBN: (纸本)9781450310536

Hardware prefetching and last-level cache (LLC) management are two independent mechanisms to mitigate the growing latency to memory. However, the interaction between LLC management and hardware prefetching has received very little attention. This paper characterizes the performance of state-of-the-art LLC management policies in the presence and absence of hardware prefetching. Although prefetching improves performance by fetching useful data in advance, it can interact with LLC management policies to introduce application performance variability. This variability stems from the fact that current replacement policies treat prefetch and demand requests identically. In order to provide better and more predictable performance, we propose Prefetch-Aware Cache Management (PACMan). PACMan dynamically estimates and mitigates the degree of prefetch-induced cache interference by modifying the cache insertion and hit promotion policies to treat demand and prefetch requests differently. Across a variety of emerging workloads, we show that PACMan eliminates the performance variability in state-of-the-art replacement policies under the influence of prefetching. In fact, PACMan improves performance consistently across multimedia, games, server, and SPEC CPU2006 workloads by an average of 21.9% over the baseline LRU policy. For multiprogrammed workloads, on a 4-core CMP, PACMan improves performance by 21.5% on average. © 2011 ACM.

关键词： multiprogramming

来源：评论

学校读者我要写书评

暂无评论

A generalization of blocking and windowing algorithms for duplicate detection

A generalization of blocking and windowing algorithms for du...

引用

2nd International Conference on Data and Knowledge Engineering, ICDKE 2011

作者： Draisbach, Uwe Naumann, Felix Hasso Plattner Institute 14482 Potsdam Germany

ISBN: (纸本)9781457708640

Duplicate detection is the process of finding multiple records in a dataset that represent the same real-world entity. Due to the enormous costs of an exhaustive comparison, typical algorithms select only promising record pairs for comparison. Two competing approaches are blocking and windowing. Blocking methods partition records into disjoint subsets, while windowing methods, in particular the Sorted Neighborhood Method, slide a window over the sorted records and compare records only within the window. We present a new algorithm called Sorted Blocks in several variants, which generalizes both approaches. To evaluate Sorted Blocks, we have conducted extensive experiments with different datasets. These show that our new algorithm needs fewer comparisons to find the same number of duplicates. © 2011 IEEE.

关键词： duplicate detection multiprogramming BLOCKING windows window Dataset numerical invariant Records as Topic

来源：评论

学校读者我要写书评

暂无评论

Performance evaluation of windowing approach on effort estimation by analogy

Performance evaluation of windowing approach on effort estim...

引用

Joint Conference of the 21st International Workshop on Software Measurement, IWSM 2011 and the 6th International Conference on Software Process and Product Measurement, MENSURA 2011

作者： Amasaki, Sousuke Takahara, Yohei Yokogawa, Tomoyuki Dept. of Systems Engineering Okayama Prefectural University Okayama Japan

ISBN: (纸本)9780769544977

Background: In effort estimation model construction, it seems effective to window training project data so that only recently finished projects are used. This is because old projects might be less representative of an organization. The past study demonstrated windowing approach works with linear regression, which is one of global models. However, this approach has not been examined with local models. Local models use subset of historical data for model construction and thus windowing approach may influence on its performance more weakly. Aim: To investigate whether windowing approach works with local models. Method: We replicated the past study with EbA. Maxwell and CSC datasets were used for an experiment. Results: Windowing approach improved predictive performance. Although the difference was insignificant in any window size, the result indicated using windowing approach has positive effect on average. Conclusions: This result contributes to understand where windowing approach works well. © 2011 IEEE.

关键词： Accuracy EbA Estimation Maintenance engineering Modeling Software Testing Training effort estimation estimation by analogy global models integrated software linear regression multiprogramming performance evaluation regression analysis software effort esti

来源：评论

学校读者我要写书评

暂无评论

Access pattern based re-reference interval table for last level cache

Access pattern based re-reference interval table for last le...

引用

2011 12th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2011

作者： Yu, Baozhong Hu, Yifan Ma, Jianliang Chen, Tianzhou College of Computer Science Zhejiang University Hangzhou China

ISBN: (纸本)9780769545646

The memory access stream to the last level cache (LLC) is a filtered version from the upper level caches. There are a large number of cache blocks with long re-reference interval in the LLC. What's worse, zero reuse blocks are in abundance in the LLC and pollute the LLC for a long time. Access pattern based re-reference interval table (RRIT), here we propose, is an efficient management technology for the LLC. RRIT introduces a table to track consecutive blocks're-reference interval. And then predicts the re-reference interval for a new incoming block. RRIT always selects the block with the max rereference interval as the victim block when a cache replacement issue occurs. Furthermore, we propose a dead block filtering (RRIT-Filter) mechanism based on RRIT. RRIT-Filter bypasses the predicted dead block and improves LLC efficiency. Our evaluations using the SPEC2006 workloads on a singlecore with a 2MB LLC show that RRIT and RRIT-Filter outperform LRU replacement on the throughput metric by an average of 6.6% and 8.6% respectively. Our evaluations with 10 multiprogrammed workloads on a 4-core CMP with an 8MB shared LLC show that our strategy outperforms LRU replacement on the throughput metric by an average of 23% and 39% respectively. © 2011 IEEE.

关键词： multiprogramming

来源：评论

学校读者我要写书评

暂无评论

Computational Efficiency of Ultrasonic Guided Wave Imaging Algorithms

引用

IEEE TRANSACTIONS ON ULTRASONICS FERROELECTRICS AND FREQUENCY CONTROL 2011年第1期58卷 244-248页

作者： Hall, James S. Michaels, Jennifer E. Georgia Inst Technol Sch Elect & Comp Engn Atlanta GA 30332 USA

Guided wave imaging techniques employed for structural health monitoring (SHM) can be computationally demanding, especially for adaptive techniques such as minimum variance distortionless response (MVDR) imaging, which requires a matrix inversion for each pixel calculation. Instantaneous windowing has been shown in previous work to improve guided wave imaging performance. The use of instantaneous windowing has the additional benefit of significantly reducing the computational requirements of image generation. This paper derives a formulation for MVDR imaging using instantaneous windowing and shows that the matrix inversion associated with MVDR imaging can be optimized, reducing the computational complexity to that of conventional delay-and-sum imaging algorithms. Additionally, a vectorized approach is presented for implementing guided wave imaging algorithms, including delay-and-sum imaging, in matrix-based software packages. The improvements in computational efficiency are quantified by measuring computation time for different array sizes, windowing assumptions, and imaging methods.

关键词： computational time Computational efficiency Imaging Image generation Software packages multiprogramming matrix inversion

来源：评论

学校读者我要写书评

暂无评论

Improving Acoustic Fall Recognition by Adaptive Signal Windowing

Improving Acoustic Fall Recognition by Adaptive Signal Windo...

引用

Annual International Conference of the IEEE Engineering in Medicine and Biology Society

作者： Yun Li Mihail Popescu K. C. Ho Daniel P. Nabelek ECE Dept. University of Missouri Health Mngmt. & Inform. Department University of Missouri Elect. and Comp. Eng. Dept. University of Missouri

ISBN: (纸本)9781424441211

Each year more than a third of elderly fall in the United States. To address this problem we are developing an acoustic fall detection system based on a microphone array. The main task of the acoustic system is to detect all the falls that occur in an indoor environment while producing as few as possible false alarms. One of the challenges of this task is to accurately locate where the fall signal comes from so that beamforming can be applied to improve the recognition of fall signals. In this paper we describe a simple fall signal location procedure that proved effective in preliminary testing.

关键词： falls Fall Elderly fall Acoustic equipment Detection systems Autumn Recognition (Psychology) indoor environment Beamforming multiprogramming Acoustics preliminary test Microphone Device Component

来源：评论

学校读者我要写书评

暂无评论

Feedback process scheduling with simple discrete-time control structures

引用

IET CONTROL THEORY AND APPLICATIONS 2010年第11期4卷 2331-2342页

作者： Leva, A. Maggio, M. Politecn Milan Dipartimento Elettron & Informaz I-20133 Milan Italy

This manuscript addresses the problem of process scheduling in a multitasking computing environment. The mainstream feedback-based approach to that problem preserves the existing scheduler, and adapts some of its parameters by means of convenient loops. On the contrary, in this research, the scheduler is entirely replaced by suitable control structures, synthesised and analysed in the discrete-time domain. The proposed approach allows for a clear interpretability of the involved parameters, whereas the complexity of the obtained scheduling solutions is comparable to the existing ones. Simulation examples support the above claims. The focus is here restricted for convenience to the preemptive single-processor case, although several generalisations are possible.

关键词： feedback process scheduling discrete-time control processor scheduling Parallel programming and algorithm theory multitasking computing feedback interpretability discrete time systems Discrete control systems multiprogramming

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：