线性局部切空间排列算法(Linear local tangent space alignment,LLTSA)是能够较好应用于模式识别问题的降维方法,但由于其属于无监督的降维方法且在降维过程中只使用全局统一的邻域参数,使得在对高维数据集进行约简时,不能利用部分样...线性局部切空间排列算法(Linear local tangent space alignment,LLTSA)是能够较好应用于模式识别问题的降维方法,但由于其属于无监督的降维方法且在降维过程中只使用全局统一的邻域参数,使得在对高维数据集进行约简时,不能利用部分样本的类别标签信息且不能根据样本空间分布的变化调整邻域参数。针对上述问题,提出了一种半监督邻域自适应线性局部切空间排列算法(Semi-supervised neighborhood self-adaptive LLTSA,SSNA-LLTSA)。该算法在LLTSA的基础上,利用部分标签信息来调整样本点与点之间的距离以形成新的距离矩阵来完成邻域构建,同时根据每个数据样本点邻域的概率密度自适应地调整邻域参数,进而得到更好的降维效果。经典的三维流形、UCI典型数据集模式识别和轴承故障诊断的实验结果表明,该算法克服了LLTSA算法无监督和使用全局统一邻域参数的不足,可更有效地寻找数据的低维本质流形,提高了识别准确率,具有一定优势。展开更多
A linear array of N mutually coupled single-mode lasers is investigated. It is shown that the intensities of N lasers are chaotically synchronized when the coupling between lasers is relatively strong. The chaotic syn...A linear array of N mutually coupled single-mode lasers is investigated. It is shown that the intensities of N lasers are chaotically synchronized when the coupling between lasers is relatively strong. The chaotic synchronization of intensities depends on the location of the lasers in the array. The chaotic synchronization appears between two outmost lasers, the second two outmost lasers, etc. There is no synchronization between nearest neighbors of the lasers. If the number of N is odd, the middle laser is never synchronized between any lasers. The chaotic synchronization of phases between nearest lasers in the array is examined by using the analytic signal and the Gaussian filter methods based on the peak of the power spectrum of the intensity. It can be seen that the message of chaotic intensity synchronization is conveyed through the phase synchronization.展开更多
文摘线性局部切空间排列算法(Linear local tangent space alignment,LLTSA)是能够较好应用于模式识别问题的降维方法,但由于其属于无监督的降维方法且在降维过程中只使用全局统一的邻域参数,使得在对高维数据集进行约简时,不能利用部分样本的类别标签信息且不能根据样本空间分布的变化调整邻域参数。针对上述问题,提出了一种半监督邻域自适应线性局部切空间排列算法(Semi-supervised neighborhood self-adaptive LLTSA,SSNA-LLTSA)。该算法在LLTSA的基础上,利用部分标签信息来调整样本点与点之间的距离以形成新的距离矩阵来完成邻域构建,同时根据每个数据样本点邻域的概率密度自适应地调整邻域参数,进而得到更好的降维效果。经典的三维流形、UCI典型数据集模式识别和轴承故障诊断的实验结果表明,该算法克服了LLTSA算法无监督和使用全局统一邻域参数的不足,可更有效地寻找数据的低维本质流形,提高了识别准确率,具有一定优势。
文摘A linear array of N mutually coupled single-mode lasers is investigated. It is shown that the intensities of N lasers are chaotically synchronized when the coupling between lasers is relatively strong. The chaotic synchronization of intensities depends on the location of the lasers in the array. The chaotic synchronization appears between two outmost lasers, the second two outmost lasers, etc. There is no synchronization between nearest neighbors of the lasers. If the number of N is odd, the middle laser is never synchronized between any lasers. The chaotic synchronization of phases between nearest lasers in the array is examined by using the analytic signal and the Gaussian filter methods based on the peak of the power spectrum of the intensity. It can be seen that the message of chaotic intensity synchronization is conveyed through the phase synchronization.