Non-negative matrix factorization (NMF) is a powerful dimension reduction method and has been widely used in many pattern recognition and computer vision problems. However, conventional NMF methods are neither robust ...
详细信息
ISBN:
(纸本)9781479919611
Non-negative matrix factorization (NMF) is a powerful dimension reduction method and has been widely used in many pattern recognition and computer vision problems. However, conventional NMF methods are neither robust enough as their loss functions are sensitive to outliers, nor discriminative because they completely ignore labels in a dataset. In this paper, we proposed a correntropy supervised NMF (CSNMF) to simultaneously overcome aforementioned deficiencies. In particular, CSNMF maximizes the correntropy between the data matrix and its reconstruction in low-dimensional space to inhibit outliers during learning the subspace, and narrows the minimizes the distances between coefficients of any two samples with the same class labels to enhance the subsequent classification performance. To solve CSNMF, we developed a multiplicative update rules and theoretically proved its convergence. Experimental results on popular face image datasets verify the effectiveness of CSNMF comparing with NMF, its supervised variants, and its robustified variants.
In smart grid, privacy implications to individuals and their family is an important issue, due to the fine-grained usage data collection. Wireless communications are considered by many utility companies to obtain info...
详细信息
Asynchrony based overlapping of computation and communication is commonly used in MPI applications. However, this overlapping introduces synchronization errors frequently in asynchronous MPI programming. In this paper...
详细信息
In this paper, we present a novel stochastic analyzing model for e2e virtualized cloud services using hierarchical Quasi-Birth Death structures (QBDs). We divide the overall virtualized cloud services into three sub-h...
详细信息
Pervasive software should be able to adapt itself to the changing environments and user requirements. Obviously, it will bring great challenges to the software engineering practice. This paper proposes AUModel, a conc...
详细信息
Multithreaded programs execute nondeterministically on conventional architectures and operating systems. This complicates many tasks, including debugging and testing. Deterministic multithreading (DMT) makes the outpu...
详细信息
Multithreaded programs execute nondeterministically on conventional architectures and operating systems. This complicates many tasks, including debugging and testing. Deterministic multithreading (DMT) makes the outpu...
详细信息
Trace-oriented runtime monitoring is a very effective method to improve the reliability of distributed systems. However, for medium-scale distributed systems, existing trace-oriented monitoring frameworks are either n...
详细信息
Trace-oriented runtime monitoring is a very effective method to improve the reliability of distributed systems. However, for medium-scale distributed systems, existing trace-oriented monitoring frameworks are either not powerful or efficient enough, or too complex and expensive to deploy and maintain. In this paper, we present MTracer, which is a lightweight trace-oriented monitoring system for medium-scale distributed systems. We have proposed and implemented several optimizations to improve the efficiency of the monitor server in MTracer. A web-based frontend is also provided to visualize a monitored system from different perspectives. We have validated MTracer in a real medium-scale environment. The results indicate that MTracer has a very lower overhead, and can handle more than 4000 events per second.
With the development of high performance computing and Web 2.0 applications,unstructured data storage becomes more and more *** RDBMS isn't efficient for big data ***,RDBMS's scalability is ***' expansion ...
详细信息
With the development of high performance computing and Web 2.0 applications,unstructured data storage becomes more and more *** RDBMS isn't efficient for big data ***,RDBMS's scalability is ***' expansion often leads to a large scale of data *** paper designs and implements a high performance distributed key-value database,which is distributed Stage *** servers are organized by a consistent hashing ring and distributed with the support of Zookeeper,a distributed service *** has a high single-node read/write *** route information is calculated by clients,which reduces the expense of expansion.
The Embarrassingly Parallel(EP) algorithm which is typical of many Monte Carloapplications provides an estimate of the upper achievable limits for double precision performance of parallel supercomputers. Recently, Int...
详细信息
The Embarrassingly Parallel(EP) algorithm which is typical of many Monte Carloapplications provides an estimate of the upper achievable limits for double precision performance of parallel supercomputers. Recently, Intel released Many Integrated Core(MIC) architecture as a many-core co-processor. MIC often offers more than 50 cores each of which can run four hardware threads as well as 512-bit vector instructions. In this paper,we describe how the EP algorithm is accelerated effectively on the platforms containing MIC using the offload execution model. The result shows that the efficientimplementation of EP algorithm on MIC can take full advantage of MIC's computational resources and achieves a speedup of 3.06 compared with that on Intel Xeon E5-2670 CPU. Based on the EP algorithm on MIC and an effective task distribution model, the implementation of EP algorithm on a CPU-MIC heterogeneous platform achieves the performance of up to2134.86 Mop/s and 4.04 times speedup compared with that on Intel Xeon E5-2670 CPU.
暂无评论