Abstract:Nonnegative tensor factorization has applications in statistics, computer vision, exploratory multiway data analysis, and blind source separation. A symmetric nonnegative tensor, which has an exact symmetric nonnegative factorization, is called a completely positive tensor. This concept extends the concept of completely positive matrices. A classical result in the theory of completely positive matrices is that a symmetric, diagonally dominated nonnegative matrix is a completely positive matrix. In this paper, we introduce strongly symmetric tensors and show that a symmetric tensor has a symmetric binary decomposition if and only if it is strongly symmetric. Then we show that a strongly symmetric, hierarchically dominated nonnegative tensor is a completely positive tensor, and present a hierarchical elimination algorithm for checking this. Numerical examples are given to illustrate this. Some other properties of completely positive tensors are discussed. In particular, we show that the completely positive tensor cone and the co-positive tensor cone of the same order are dual to each other.

A Semidefinite Algorithm for Completely Positive Tensor Decomposition.

Completely positive tensor decomposition

Completely Positive Tensors in the Complex Field

Nonnegative Tensor Factorization, Completely Positive Tensors, and a Hierarchical Elimination Algorithm.

A Semidefinite Relaxation Algorithm for Checking Completely Positive Separable Matrices

Completely Positive Binary Tensors

A Hierarchy of Semidefinite Relaxations for Completely Positive Tensor Optimization Problems.

A COMPLETE SEMIDEFINITE ALGORITHM FOR DETECTING COPOSITIVE MATRICES AND TENSORS

Nonnegative Tensor Factorization, Completely Positive Tensors and an Hierarchical Elimination Algorithm

Dehomogenization for completely positive tensors

A Semidefinite Method for Tensor Complementarity Problems

Computational complexity of decomposing a symmetric matrix as a sum of positive semidefinite and diagonal matrices

Successive Partial-Symmetric Rank-One Algorithms for Almost Unitarily Decomposable Conjugate Partial-Symmetric Tensors

On Semi-Definiteness and Minimal H-eigenvalue of a Symmetric Space Tensor Using Nonnegative Polynomial Optimization Techniques.

Hermitian Completely Positive Matrices

The decompositions and positive semidefiniteness of fourth-order conjugate partial-symmetric tensors with applications

A Direct Proof of Uniqueness of Square-Root of a Positive Semi-Definite Tensor

Semidefinite Relaxation Methods for Tensor Absolute Value Equations.

Three Dimensional Strongly Symmetric Circulant Tensors

Monotonically Positive Matrices

Partially Positive Matrices