【作者】 刘丽娜； 温加睿； 马世伟；

【Author】 LIU Lina;WEN Jiarui;MA Shiwei;School of Mechatronic Engineering and Automation, Shanghai University;School of Electrical and Electronic Engineering,Shandong University of technology;

【机构】 上海大学机电工程与自动化学院； 电气与电子工程学院,山东理工大学；

【摘要】 针对目前已有的非线性降维算法存在计算复杂性高,难以处理大型数据集和增量化降维问题,本文提出了一种基于局部约束字典学习的非线性降维算法。该方法通过重构一些潜在标志点的局部内在流形,并在数据处理过程中将训练数据和未知数据一起嵌入到内在流形中,使得数据的内在几何结构特征得以保持。与已有非线性降维方法相比,该算法具有计算复杂度低、存储空间小和通用性强的特点,可以很好的解决增量化降维问题,易于处理大型数据集。另外,该算法也可以解决高维数据的重构问题,与已有重构方法相比具有计算简单,重构误差较低的特点。实验结果表明了算法的有效性。更多还原

【Abstract】 In order to solve the problems of current existing nonlinear dimensionality reduction algorithms, such as its high computational complexity, its difficulty to deal with large scale data sets and out-of-sample extension problem, this paper proposes a nonlinear dimensionality reduction algorithm which is based on locality constrained dictionary learning. Through reconstructing the local intrinsic manifold of some potential landmarks and embedding the training datasets and unknown datasets into the intrinsic manifold, the intrinsic local geometric construction feature of the datasets are maintained. Compared with the existing methods, it has the characteristics of lower computational complexity, smaller storage space and strong generality. It can be used to solve the out of sample extension and large-scale data sets problem. In addition, this algorithm can also be used to deal with the high dimensional data reconstruction problem. And it has the characteristics of simple calculation and lower reconstruction error. The experimental results have proven the efficiency of the proposed algorithm.更多还原