RI-Small: Approximation Based Reasoning and Planning

RI-Small：基于近似的推理和规划

• 批准号: 0812267
• 负责人: Son Tran
• 金额: $ 33.99万
• 依托单位: New Mexico State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2008
• 资助国家: 美国
• 起止时间: 2008-09-01 至 2012-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0812267&HistoricalAwards=false
• 关键词: RI; Small; Approximation; Based; Reasoning

The project is developing a framework for efficient reasoning about effects of actions in temporal action theories (TAT) with incomplete information and state constraints in the form of static causal laws. To cope with the high complexity of temporal reasoning and planning problems, the project develops and exploits an approximation theory for query answering and planning in TAT. This theory is being used as the basis for sound and efficient algorithms for hypothetical reasoning and planning.This framework guides the development of state-of-the-art temporal conformant planners, robust in the face of a variety of contingencies, which can deal with various classes of preferences. Techniques developed will application, for example, in the composition of web services for comparative data analysis in evolutionary biology.

该项目正在开发一个框架，用于在不完全信息和静态因果律形式的状态约束下，对时间行动理论（达特）中的行动效果进行有效推理。为了科普时间推理和规划问题的高度复杂性，该项目开发和利用近似理论的查询回答和规划在达特。这一理论被用来作为假设推理和规划的合理和有效的算法的基础。这一框架指导了最先进的时间一致规划者的发展，在面对各种突发事件时具有鲁棒性，可以处理各种类别的偏好。开发的技术将应用于，例如，在进化生物学的比较数据分析的Web服务的组成。