Completely Positive Graphs and the SPN Completion Problem

来源 :2016年张量和矩阵学术研讨会(International conference on Tensor, Matrix a | 被引量 : 0次 | 上传用户:rtpy1015
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
  A graph G is completely Positive if every doubly nonnegative matrix realization of G is completely positive. A matrix is SPN if it is the sum of a PSD matrix and a symmetric nonnegative matrix. Let C be a class of square matrices with the property that for every matrix A in C, the principal submatrices of A are also in C. For a graph G with vertices 1, 2, , n a G-partial matrix is an nxn partial matrix whose ij entry is specified iff ij is an edge of G or i=j. The C completion problem is the question of characterizing the graphs G for which every G-partial C matrix can be completed to a matrix in C. In the talk we will show how the solution of the SPN completion problem is related to the characterization of completely positive graphs.
其他文献
  We generalize the matrix Kronecker product to tensors and propose the tensor Kro-necker product singular value decomposition (TKPSVD) that decomposes a real
会议
会议
会议
  Recently, the study on tensor eigenvalue complementarity problem (TEiCP) has at-tracted much attention due to its various applications in polynomial optimiz
会议
会议
会议
会议
会议
  The spectral theory of higher-order symmetric tensors is an important tool to reveal some important properties of a hypergraph via its adjacency tensor, Lap
会议
会议