ID 原文 译文
7934 此外,由于增广三线性模型为实数,因而所提算法的计算复杂度低。 In addition, because of the augmented three linear model for the real number, thus the proposed algorithm has low computing complexity.
7935 最后,通过仿真实验验证了所提算法的有效性。 Finally, the effectiveness of the proposed algorithm is verified by simulation experiments.
7936 二维稀疏信号的重构可以通过解多观测向量的稀疏表示问题来实现。 Two-dimensional sparse signal reconstruction can be problem by solving the sparse representation of the observation vector.
7937 然而,当各向量的稀疏结构不同时,将稀疏恢复算法拓展到多观测向量模型的方法将不再有效。 However, when the vector sparsity structure is not at the same time, will sparse recovery algorithm to expand to more observation vector model methods will no longer valid.
7938 提出了一种序列降采样重构的方法用于实现稀疏矩阵的重构。 This paper proposes a sequence down sampling to refactor methods used to achieve sparse matrix.
7939 该方法通过构造降采样矩阵,大幅降低稀疏矩阵信号的稀疏度,再通过多观测向量序列观测和恢复,完成对稀疏矩阵的重构。 By constructing the sampling matrix, this method greatly reduce the sparse degree of sparse matrix signal, again through the observation vector sequence observation and recovery, to complete the reconstruction of sparse matrix.
7940 理论分析表明,所提方法能够实现对高稀疏度矩阵的高概率重构。 Theoretical analysis shows that the proposed method can achieve high probability for a high degree of sparse matrix remodeling.
7941 实验表明,所提算法能够有效地实现二维稀疏信号和图像重构。 Experiments show that the proposed algorithm can effectively realize 2-d sparse signal and image reconstruction.
7942 针对传统模糊认知图学习算法仅优化概念结点间有向弧的权值误差而造成模型拟合准确程度不高的问题,将多目标演化的思想用于模糊认知图学习算法,设计了以有向弧权值误差与误差权重同时最小化为目标的多目标模糊认知图学习模型,降低了学习算法对权值的依赖。 Learning algorithm in view of the traditional fuzzy cognitive map (only optimize the concept of directed arcs between nodes weight error caused by model fitting accuracy is not high, the multi-objective evolutionary thought learning algorithm for fuzzy cognitive map, designed to have the arc weights of error and error weighting targets at the same time minimizing learning model of multi-objective fuzzy cognitive map, reduced the dependence on weight learning algorithm.
7943 为有效求解该多目标优化模型,提出了一种基于坐标变换的多目标演化算法,分析了算法的参数设置方法与计算复杂度。 To effectively solve the multi-objective optimization model, this paper puts forward a multi-objective evolutionary algorithm based on coordinate transform, the parameter setting method and the computational complexity of the algorithm is analyzed.