A new approach for rules-based optical proximity correction is presented.The discussion addresses on how to select and construct more concise and practical rules-base as well as how to apply that rules-base.Based on t...A new approach for rules-based optical proximity correction is presented.The discussion addresses on how to select and construct more concise and practical rules-base as well as how to apply that rules-base.Based on those ideas,several primary rules are suggested.The v-support vector regression method is used to generate a mathematical expression according to rule data.It enables to make correction according to any given rules parameters.Experimental results demonstrate applying rules calculated from the expression match well with that from the rule table.展开更多
针对现有模式分类方法不能较好地保持数据空间的局部流形信息或差异信息等问题,提出一种基于流形学习的局部保留最大信息差v-支持向量机(Locality-preserved maximum information variance v-support vector machine,v-LPMIVSVM).对于...针对现有模式分类方法不能较好地保持数据空间的局部流形信息或差异信息等问题,提出一种基于流形学习的局部保留最大信息差v-支持向量机(Locality-preserved maximum information variance v-support vector machine,v-LPMIVSVM).对于模式分类问题,v-LPMIVSVM引入局部同类离散度和局部异类离散度概念,分别体现输入空间局部流形结构和局部差异(或判别)信息,通过最小化局部同类离散度和最大化局部异类离散度,优化分类器的投影方向.同时,v-LPMIVSVM采用适于流形数据的测地线距离来度量数据点对间的相似性,以更好地反映流形数据的本质结构.人造和实际数据集实验结果显示所提方法具有良好的泛化性能.展开更多
文摘A new approach for rules-based optical proximity correction is presented.The discussion addresses on how to select and construct more concise and practical rules-base as well as how to apply that rules-base.Based on those ideas,several primary rules are suggested.The v-support vector regression method is used to generate a mathematical expression according to rule data.It enables to make correction according to any given rules parameters.Experimental results demonstrate applying rules calculated from the expression match well with that from the rule table.
文摘针对现有模式分类方法不能较好地保持数据空间的局部流形信息或差异信息等问题,提出一种基于流形学习的局部保留最大信息差v-支持向量机(Locality-preserved maximum information variance v-support vector machine,v-LPMIVSVM).对于模式分类问题,v-LPMIVSVM引入局部同类离散度和局部异类离散度概念,分别体现输入空间局部流形结构和局部差异(或判别)信息,通过最小化局部同类离散度和最大化局部异类离散度,优化分类器的投影方向.同时,v-LPMIVSVM采用适于流形数据的测地线距离来度量数据点对间的相似性,以更好地反映流形数据的本质结构.人造和实际数据集实验结果显示所提方法具有良好的泛化性能.
