Entrywise Approximation for Matrix Inversion and Linear Systems
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915568582393856 |
|---|---|
| author | Ghadiri, Mehrdad Nguyen, Hoai-An Yang, Junzhao |
| author_facet | Ghadiri, Mehrdad Nguyen, Hoai-An Yang, Junzhao |
| contents | We study the bit complexity of inverting diagonally dominant matrices, which are associated with random walk quantities such as hitting times and escape probabilities. Such quantities can be exponentially small, even on undirected unit-weighted graphs. However, their nonnegativity suggests that they can be approximated entrywise, leading to a stronger notion of approximation than vector norm-based error.
Under this notion of error, existing Laplacian solvers and fast matrix multiplication approaches have bit complexities of $mn^2$ and $n^{ω+1}$, respectively, where $m$ is the number of nonzero entries in the matrix, $n$ is its size, and $ω$ is the matrix multiplication exponent.
We present algorithms that compute entrywise $\exp(ε)$-approximate inverses of row diagonally dominant $L$-matrices (RDDL) in two settings: (1) when the matrix entries are given in floating-point representation; (2) when they are given in fixed-point representation.
For floating-point inputs, we present a cubic-time algorithm and show that it has an optimal running time under the all-pairs shortest paths (APSP) conjecture.
For fixed-point inputs, we present several algorithms for solving linear systems and inverting RDDL and SDDM matrices, all with high probability.
Omitting logarithmic factors:
(1) For SDDM matrices, we provide an algorithm for solving a linear system with entrywise approximation guarantees using $\tilde{O}(m\sqrt{n})$ bit operations, and another for computing an entrywise approximate inverse using $\tilde{O}(mn)$ bit operations.
(2) For RDDL matrices, we present an algorithm for solving a linear system using $\tilde{O}(mn^{1+o(1)})$ bit operations, and two algorithms for computing an entrywise approximate inverse: one using $\tilde{O}(n^{ω+0.5})$ bit operations, and the other using $\tilde{O}(mn^{1.5+o(1)})$ bit operations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_19054 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Entrywise Approximation for Matrix Inversion and Linear Systems Ghadiri, Mehrdad Nguyen, Hoai-An Yang, Junzhao Data Structures and Algorithms Numerical Analysis 65Y20, 65F10, 65F05, 15A09 G.1.3; F.2.1 We study the bit complexity of inverting diagonally dominant matrices, which are associated with random walk quantities such as hitting times and escape probabilities. Such quantities can be exponentially small, even on undirected unit-weighted graphs. However, their nonnegativity suggests that they can be approximated entrywise, leading to a stronger notion of approximation than vector norm-based error. Under this notion of error, existing Laplacian solvers and fast matrix multiplication approaches have bit complexities of $mn^2$ and $n^{ω+1}$, respectively, where $m$ is the number of nonzero entries in the matrix, $n$ is its size, and $ω$ is the matrix multiplication exponent. We present algorithms that compute entrywise $\exp(ε)$-approximate inverses of row diagonally dominant $L$-matrices (RDDL) in two settings: (1) when the matrix entries are given in floating-point representation; (2) when they are given in fixed-point representation. For floating-point inputs, we present a cubic-time algorithm and show that it has an optimal running time under the all-pairs shortest paths (APSP) conjecture. For fixed-point inputs, we present several algorithms for solving linear systems and inverting RDDL and SDDM matrices, all with high probability. Omitting logarithmic factors: (1) For SDDM matrices, we provide an algorithm for solving a linear system with entrywise approximation guarantees using $\tilde{O}(m\sqrt{n})$ bit operations, and another for computing an entrywise approximate inverse using $\tilde{O}(mn)$ bit operations. (2) For RDDL matrices, we present an algorithm for solving a linear system using $\tilde{O}(mn^{1+o(1)})$ bit operations, and two algorithms for computing an entrywise approximate inverse: one using $\tilde{O}(n^{ω+0.5})$ bit operations, and the other using $\tilde{O}(mn^{1.5+o(1)})$ bit operations. |
| title | Entrywise Approximation for Matrix Inversion and Linear Systems |
| topic | Data Structures and Algorithms Numerical Analysis 65Y20, 65F10, 65F05, 15A09 G.1.3; F.2.1 |
| url | https://arxiv.org/abs/2504.19054 |