A fast and efficient Newton-Shultz-type iterative method is presented to compute the inverse of an invertible tensor. Analysis of the convergence error shows that the proposed method has the sixth order convergence. It is shown that the proposed algorithm can be used for finding the Moore-Penrose inverse of tensors. Computational complexities of the algorithm is presented to support the theoretical aspects of the paper. Using the new method, we obtain a new preconditioner to solve the multilinear system $\mathcal{A}\ast_N\mathcal{X}=\mathcal{B}$. The effectiveness and accuracy of this method are re-verified by several numerical examples. Finally, some conclusions are given.
Khosravi Dehdezi, E., & Karimi, S. (2021). A fast and efficient Newton-Shultz-type iterative method for computing inverse and Moore-Penrose inverse of tensors. Journal of Mathematical Modeling, 9(4), 645-664. doi: 10.22124/jmm.2021.19005.1627
MLA
Eisa Khosravi Dehdezi; Saeed Karimi. "A fast and efficient Newton-Shultz-type iterative method for computing inverse and Moore-Penrose inverse of tensors". Journal of Mathematical Modeling, 9, 4, 2021, 645-664. doi: 10.22124/jmm.2021.19005.1627
HARVARD
Khosravi Dehdezi, E., Karimi, S. (2021). 'A fast and efficient Newton-Shultz-type iterative method for computing inverse and Moore-Penrose inverse of tensors', Journal of Mathematical Modeling, 9(4), pp. 645-664. doi: 10.22124/jmm.2021.19005.1627
VANCOUVER
Khosravi Dehdezi, E., Karimi, S. A fast and efficient Newton-Shultz-type iterative method for computing inverse and Moore-Penrose inverse of tensors. Journal of Mathematical Modeling, 2021; 9(4): 645-664. doi: 10.22124/jmm.2021.19005.1627
)