Detecting element failures is a relevant issue in distributed systems. A fault tolerant system needs to detect a failure and recover from it promptly. In fact, traditional approaches to fault tolerance are usually not...
详细信息
ISBN:
(纸本)9788988678251
Detecting element failures is a relevant issue in distributed systems. A fault tolerant system needs to detect a failure and recover from it promptly. In fact, traditional approaches to fault tolerance are usually not completely free from errors during the failure detection phase;a good failure detector is thus a very important component of them to minimize these errors. In this paper we present a failure detector able to monitor both asynchronous and synchronous elements of a distributed system by exchanging messages withthe monitored elements. In order to assess the health status of monitored elements our failure detector relies on a simple query/ACK mechanism, which however requires a reliable timeout estimate in order to properly set the monitoring interval. To this purpose our failure detector uses the history of past estimates to compute new values for both quantities. the model proposed here introduces a new label to tag monitored elements, besides those used in traditional failures detectors. To evaluate this work, we compared it with two other algorithms by computing performance metrics, such as specificity and sensitivity, and by considering the number of required control packets. We also compared the performance of the failure detectors by computingtheir detection time.
We present the first temporal-difference learning algorithm for off-policy control with unrestricted linear function approximation whose per-time-step complexity is linear in the number of features. Our algorithm, Gre...
详细信息
ISBN:
(纸本)9781605589077
We present the first temporal-difference learning algorithm for off-policy control with unrestricted linear function approximation whose per-time-step complexity is linear in the number of features. Our algorithm, Greedy- GQ, is an extension of recent work on gradient temporal-difference learning, which has hitherto been restricted to a prediction (policy evaluation) setting, to a control setting in which the target policy is greedy with respect to a linear approximation to the optimal action-value function. A limitation of our control setting is that we require the behavior policy to be stationary. We call this setting latent learning because the optimal policy, though learned, is not manifest in behavior. Popular off-policy algorithms such as Q-learning are known to be unstable in this setting when used with linear function ap-proximation. Copyright 2010 by the author(s)/owner(s).
the proceedings contain 46 papers. the topics discussed include: customer co-design of computer mouse for mass customization without causing mass confusion;assembly time modeling through connective complexity metrics;...
ISBN:
(纸本)9780769542935
the proceedings contain 46 papers. the topics discussed include: customer co-design of computer mouse for mass customization without causing mass confusion;assembly time modeling through connective complexity metrics;green product development by using life cycle assessment (LCA), theory of inventive of problems solving (TRIZ);environmentally-friendly nano-fluid minimum quantity lubrication (MQL) meso-scale grinding process using nano-diamond particles;fabrication of electronics devices with multi-material drop-on-demand dispensing system;the size matching and scaling method: a synthesis method for the design of mesoscale cellular structures;study on fruit quality inspection based on its surface color in produce logistics;GPU-based optimization of tool path planning in 5-axis flank milling;and the research of optimal algorithm for task scheduling underground wireless network based on distributed computing.
Face recognition has a great demands in human authentication and it becomes one of the most intensive field of biometrics research areas. In this paper, we present a bio-inspired face recognition system based on linea...
详细信息
Successful deployment of new network protocols on the Future Internet is not a trivial task. Deployable protocol design is necessary but not sufficient condition for protocol's success, unless it takes all stakeho...
详细信息
ISBN:
(纸本)9781424481668
Successful deployment of new network protocols on the Future Internet is not a trivial task. Deployable protocol design is necessary but not sufficient condition for protocol's success, unless it takes all stakeholders involved in the deployment process into account. this paper investigates the challenges of deploying a new transport protocol on the Internet, using Multipath TCP - a TCP variant that transmits along several network paths at the same time - as an example and proposes a framework for its adoption process based on diffusion theory. the paper distinguishes the roles of adopters and other stakeholders in the deployment process, and presents scenarios that enhance Multipath TCP deployment and adoption. One key finding is that the role of end users is not of significant importance for Multipath TCP deployment, because they are not necessarily in a position to make a conscious adoption decision.
Spatial Scene Similarity Assessment (SSSA) is an essential problem in spatial analysis, spatial query, and map generalization, etc. In SSSA, spatial scene similarity needs to be compared between query spatial scene an...
详细信息
ISBN:
(纸本)9781450304320
Spatial Scene Similarity Assessment (SSSA) is an essential problem in spatial analysis, spatial query, and map generalization, etc. In SSSA, spatial scene similarity needs to be compared between query spatial scene and each candidate spatial scene. the computational complexity of spatial scene comparison often cannot be resolved by sequential computing model. In this paper, we analyze the computational complexity of SSSA and develop a parallel processing method and associated algorithms for SSSA based on Hadoop. the COOT (Cell Object Overlay Times) is proposed as a data locality strategy. the experiment results demonstrate that MapReduce on Hadoop significantly improve SSSA in computing performance and data processing capability. Copyright 2010 ACM.
Task scheduling is one of the most prominent problems in the era of parallel computing. We find scheduling algorithms in every domain of computer science, e.g., mapping multiprocessor tasks to clusters, mapping jobs t...
详细信息
the proceedings contain 210 papers. the special focus in this conference is on Artificial Neural Networks. the topics include: Feature-Preserving 3D thumbnail creation with voxel-based two-phase decomposition;learning...
ISBN:
(纸本)9783642172731
the proceedings contain 210 papers. the special focus in this conference is on Artificial Neural Networks. the topics include: Feature-Preserving 3D thumbnail creation with voxel-based two-phase decomposition;learning scene entries and exits using coherent motion regions;adding facial actions into 3D model search to analyse behaviour in an unconstrained environment;aggregating low-level features for human action recognition;Incorporating social entropy for crowd behavior detection using SVM;introducing a statistical behavior model into camera-based fall detection;on contrast-preserving visualisation of multispectral datasets;color gamut extension by projector-camera system;shading attenuation in human skin color images;attribute-filtering and knowledge extraction for vessel segmentation;color constancy algorithms for object and face recognition;chromatic sensitivity of illumination change compensation techniques;study on image color stealing in log-polar space;How to overcome perceptual aliasing in ASIFT?;Speeding up HOG and LBP features for pedestrian detection by multiresolution techniques;Utilizing invariant descriptors for finger spelling american sign language using SVM;Bivariate feature localization for SIFT assuming a gaussian feature shape;linear dimensionality reduction through eigenvector selection for object recognition;symmetry enhanced adaboost;object category classification using occluding contours;a human inspired local ratio-based algorithm for edge detection in fluorescent cell images;fractal map: Fractal-based 2D expansion method for multi-scale high-dimensional data visualization;visual network analysis of dynamic metabolic pathways;interpolating 3D diffusion tensors in 2D planar domain by locating degenerate lines;indented pixel tree plots;visualizing multivariate hierarchic data using enhanced radial space-filling layout;attention-based target localization using multiple instance learning;lattice-boltzmann water waves.
Recently, we have proposed a recursive partitioning based layout for multi-core computations on sparse matrices. Based on positive results of our initial experiments with matrixvector multiplication, we discuss how th...
详细信息
the proceedings contain 210 papers. the special focus in this conference is on Artificial Neural Networks. the topics include: Feature-Preserving 3D thumbnail creation with voxel-based two-phase decomposition;learning...
ISBN:
(纸本)9783642172885
the proceedings contain 210 papers. the special focus in this conference is on Artificial Neural Networks. the topics include: Feature-Preserving 3D thumbnail creation with voxel-based two-phase decomposition;learning scene entries and exits using coherent motion regions;adding facial actions into 3D model search to analyse behaviour in an unconstrained environment;aggregating low-level features for human action recognition;Incorporating social entropy for crowd behavior detection using SVM;introducing a statistical behavior model into camera-based fall detection;on contrast-preserving visualisation of multispectral datasets;color gamut extension by projector-camera system;shading attenuation in human skin color images;attribute-filtering and knowledge extraction for vessel segmentation;color constancy algorithms for object and face recognition;chromatic sensitivity of illumination change compensation techniques;study on image color stealing in log-polar space;How to overcome perceptual aliasing in ASIFT?;Speeding up HOG and LBP features for pedestrian detection by multiresolution techniques;Utilizing invariant descriptors for finger spelling american sign language using SVM;Bivariate feature localization for SIFT assuming a gaussian feature shape;linear dimensionality reduction through eigenvector selection for object recognition;symmetry enhanced adaboost;object category classification using occluding contours;a human inspired local ratio-based algorithm for edge detection in fluorescent cell images;fractal map: Fractal-based 2D expansion method for multi-scale high-dimensional data visualization;visual network analysis of dynamic metabolic pathways;interpolating 3D diffusion tensors in 2D planar domain by locating degenerate lines;indented pixel tree plots;visualizing multivariate hierarchic data using enhanced radial space-filling layout;attention-based target localization using multiple instance learning;lattice-boltzmann water waves.
暂无评论