Inexact Laplacian Solver (SDD Systems Solvers)

From Algorithm Wiki
Jump to navigation Jump to search

Description

This problem refers to solving equations of the form $Lx = b$ where $L$ is a Laplacian of a graph. In other words, this is solving equations of the form $Ax = b$ for a SDD matrix $A$.

This variation of the problem permits some error.

Related Problems

Related: Exact Laplacian Solver

Parameters

$n$: dimension of matrix

Table of Algorithms

Name Year Time Space Approximation Factor Model Reference
Spielman, Teng 2004 $O(m \log^c n)$ $O(n)$ \epsilon Deterministic Time
Vaidya 1990 $O(mn^{({3}/{4})$}) $O(n)$ \epsilon Deterministic
Gremban; Miller; Zagha 1995 $O(n^{2})$ $O(n^{2})$ ? Deterministic Time & Space
Bern; Gilbert; Hendrickson 2006 $O(n \log \log n)$ Exact Deterministic Time
Boman; Hendrickson 2003 $\tilde{O}(mn^{({1}/{2})})$ \epsilon Deterministic Time
Boman; Chen; Hendrickson; Toledo 2004 $O(n \log({1}/ϵ)$ ) $O(n)$ \epsilon Deterministic Time
Koutis; Miller and Peng 2010 $\tilde{O}(m \log n)$ $O(n)$ \epsilon Deterministic Time
Koutis; Miller and Peng 2011 $\tilde{O}(m \log n)$ $O(n)$ \epsilon Deterministic Time
Blelloch; Koutis; Miller; Tangwongsan 2010 $O(n \log n)$ m + $O(n/\log n)$ ? Deterministic Time & Space
Briggs; Henson; McCormick 2000 $O(n^{1.{2}5} \log \log n)$ Exact Deterministic Time
Kelner; Orecchia; Sidford; Zhu 2013 $O(m \log^{2} (n)$ \log \log n) $O(n)$ \epsilon Deterministic Time
Lee; Peng; Spielman 2015 $O(n)$ $O(n)$ The sparse Cholesky decomposition is a 2-approximation Deterministic Time
Daitch; Spielman 2007 $O(n^{5/4} (\log^{2} (n)$ \log\log n)^{3/4} \log({1}/ϵ)) $O(n)$ \epsilon Deterministic Time

Time Complexity Graph

SDD Systems Solvers - Inexact Laplacian Solver - Time.png