CISE Research Instrumentation: Instrumentation for Experimental Research in Machine Learning, Molecular Dynamics, Probabilistic Reasoning, and Software Maintenance

9818414Dietterich, Thomas G.Rudd, Walter G.Oregon State university机器学习，分子动力学，概率推理和软件维护研究仪器该研究仪器支持以下研究项目：-实验机器学习研究，-时间维度加速模拟，-动态环境中的概率方法，以及-基于程序分析的系统维护和测试技术的实验。为了支持上述项目，该奖项将用于俄勒冈州立大学计算机科学系购买双处理器Pentium II计算机和相关网络硬件。该设备将支持四个项目的实验研究。在第一个项目中，将为图像分析和语言处理等复杂问题开发机器学习工具。这些工具将有助于更快、更准确地解决相关问题。第二个项目集中于为复杂粒子模拟开发更快的算法。这些算法将使理解和预测化合物在制造（例如粘合剂、飞机部件）和环境清理（例如土壤中的污染物流动）中的行为成为可能。第三个项目研究不确定条件下推理的改进技术。这些技术允许构建更快、更准确的系统，用于汽车、航天器和工厂等复杂系统的实时诊断和修复。最后一个项目集中于开发和应用程序分析的新方法和工具，以改进复杂软件系统的测试和维护。这些工具有望降低成本，提高新的和现有软件系统的可靠性。

9818414Dietterich, Thomas G.Rudd, Walter G.Oregon State UniversityInstrumentation for Research in Machine Learning, Molecular Dynamics, Probabilistic Reasoning, and Software MaintenanceThis research instrumentation enables research projects in:- Experimental Machine Learning Research,- Accelerating Simulations in the Time Dimension,- Probabilistic Methods in Dynamic Environments, and - Experimentation with Program-Analysis-Based Systems Maintenance and Testing Techniques.To support the aforementioned projects, this award contributes to the purchase of dual-processor Pentium II computers and associated networking hardware by the Department of Computer Science at Oregon State University. The equipment will support experimental research on four projects. In the first project, machine learning tools will be developed for complex problems such as image analysis and language processing. These tools will help solve related problems faster and with higher accuracy. The second project concentrates in developing faster algorithms for complex particle simulations. These algorithms will make it possible to understand and predict the behavior of chemical compounds in manufacturing (e.g., adhesives, aircraft parts) and environmental cleanup (e.g. contaminant flow in soils). The third project studies improved techniques for reasoning under uncertainty. These techniques allow building faster and more accurate systems for real-time diagnosis and repair of complex systems such as automobiles, spacecraft, and factories. The last project concentrates in developing and applying new methods and tools for program analysis to improve the testing and maintenance of complex software systems. These tools are expected to reduce the cost and improve the reliability of new and existing software systems.