Scientific Computing and Numerics (SCAN) Seminar

Arvind SaibabaNorth Carolina State University
Structured matrix approximations via tensor decompositions

Monday, October 4, 2021 - 1:30pm
Gates G13

We provide a computational framework for approximating a class of structured matrices (e.g., block Toeplitz, block banded). Our approach has three steps: map the structured matrix to tensors, use tensor compression algorithms, and map the compressed tensors back to obtain two different matrix representations — sum of Kronecker products and block low-rank format. The use of tensor decompositions enable us to uncover latent structure in the matrices and lead to computationally efficient algorithms. The resulting matrix approximations are memory efficient, easy to compute with, and preserve the error due to the tensor compression in the Frobenius norm. While our framework is quite general, we illustrate the potential of our method on structured matrices from three applications: system identification, space-time covariance matrices, and test matrices from SuiteSparse collection. See preprint: https://arxiv.org/abs/2105.01170.