Home About us Contact
|
Home About us Contact | ||
|
|
||
Complex Algorithms (complex + algorithms)
Selected AbstractsEvaluation Metrics in Classification: A Quantification of Distance-BiasCOMPUTATIONAL INTELLIGENCE, Issue 3 2003Ricardo Vilalta This article provides a characterization of bias for evaluation metrics in classification (e.g., Information Gain, Gini, ,2, etc.). Our characterization provides a uniform representation for all traditional evaluation metrics. Such representation leads naturally to a measure for the distance between the bias of two evaluation metrics. We give a practical value to our measure by observing the distance between the bias of two evaluation metrics and its correlation with differences in predictive accuracy when we compare two versions of the same learning algorithm that differ in the evaluation metric only. Experiments on real-world domains show how the expectations on accuracy differences generated by the distance-bias measure correlate with actual differences when the learning algorithm is simple (e.g., search for the best single feature or the best single rule). The correlation, however, weakens with more complex algorithms (e.g., learning decision trees). Our results show how interaction among learning components is a key factor to understand learning performance. [source] In search of simplicity: a self-organizing group communication overlayCONCURRENCY AND COMPUTATION: PRACTICE & EXPERIENCE, Issue 7 2010Matei Ripeanu Abstract Group communication primitives have broad utility as building blocks for distributed applications. The challenge is to create and maintain the distributed structures that support these primitives while accounting for volatile end-nodes and variable network characteristics. Most solutions proposed to date rely on complex algorithms or on global information, thus limiting the scale of deployments and acceptance outside the academic realm. This article introduces a low-complexity, self-organizing solution for building and maintaining data dissemination trees, which we refer to as Unstructured Multi-source Overlay (UMO). UMO uses traditional distributed systems techniques: layering, soft-state, and passive data collection to adapt to the dynamics of the physical network and maintain data dissemination trees. The result is a simple, adaptive system with lower overheads than more complex alternatives. We implemented UMO and evaluated it on a 100-node PlanetLab testbed and on up to 1024-node emulated ModelNet networks. Extensive experimental evaluations demonstrate UMOs low overhead, efficient network usage compared with alternative solutions, and the ability to quickly adapt to network changes and to recover from failures. Copyright © 2009 John Wiley & Sons, Ltd. [source] High-confidence control: Ensuring reliability in high-performance real-time systemsINTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS, Issue 4 2004Tariq Samad Technology transfer is an especially difficult proposition for real-time control. To facilitate it, we need to complement the "high-performance" orientation of control research with an emphasis on establishing "high confidence" in real-time implementation. Two particular problems are discussed and recent research directed at their solutions is presented. First, the use of anytime algorithms requires dynamic resource management technology that generally is not available today in real-time systems. Second, complex algorithms have unpredictable computational characteristics that, nevertheless, need to be modeled; statistical verification is suggested as a possible approach. In both cases, a synthesis of control engineering and computer science is required if effective solutions are to be devised. Simulation-based demonstrations with uninhabited aerial vehicles (UAVs) serve to illustrate the research efforts. © 2004 Wiley Periodicals, Inc. [source] Transient elastography and other noninvasive tests to assess hepatic fibrosis in patients with viral hepatitisJOURNAL OF VIRAL HEPATITIS, Issue 5 2009Laurent Castera Summary., The limitations of liver biopsy (invasive procedure, sampling errors, inter-observer variability and nondynamic fibrosis evaluation) have stimulated the search for noninvasive approaches for the assessment of liver fibrosis in patients with viral hepatitis. A variety of methods including the measurement of liver stiffness, using transient elastography, and serum markers, ranging from routine laboratory tests to more complex algorithms or indices combining the results of panels of markers, have been proposed. Among serum indices, Fibrotest has been the most extensively studied and validated. Transient elastography appears as a promising method but has been mostly validated in chronic hepatitis C with performance equivalent to that of serum markers for the diagnosis of significant fibrosis. The combination of both approaches as first-line assessment of liver fibrosis could avoid the performance of liver biopsy in the majority of patients with chronic hepatitis C, a strategy that deserves further evaluation in patients with hepatitis B or HIV-HCV coinfection. Transient elastography also appears to be an excellent tool for early detection of cirrhosis and may have prognostic value in this setting. Guidelines are now awaited for the use of noninvasive methods in clinical practice. [source] | ||||||||||