Positive Definite, Hermitian Matrix: Difference between revisions

From Algorithm Wiki
Jump to navigation Jump to search
No edit summary
No edit summary
Line 40: Line 40:
[[File:Linear System - Positive Definite, Hermitian Matrix - Space.png|1000px]]
[[File:Linear System - Positive Definite, Hermitian Matrix - Space.png|1000px]]


== Pareto Frontier Improvements Graph ==  
== Space-Time Tradeoff Improvements ==  


[[File:Linear System - Positive Definite, Hermitian Matrix - Pareto Frontier.png|1000px]]
[[File:Linear System - Positive Definite, Hermitian Matrix - Pareto Frontier.png|1000px]]

Revision as of 14:34, 15 February 2023

Description

In this case, we restrict $A$ to be positive definite and hermitian (or symmetric, if $A$ is real-valued).

Related Problems

Generalizations: General Linear System

Related: Sparse Linear System, Non-Definite, Symmetric Matrix, Toeplitz Matrix, Vandermonde Matrix

Parameters

n: number of variables and number of equations

m: number of nonzero entries in matrix

k: ratio between largest and smallest eigenvalues

Table of Algorithms

Name Year Time Space Approximation Factor Model Reference
Gaussian-Jordan Elimination -150 $O(n^{3})$ $O(n^{2})$ Exact Deterministic
Cholesky 1940 $O(n^{3})$ $O(n^{2})$ Exact Deterministic

Time Complexity Graph

Linear System - Positive Definite, Hermitian Matrix - Time.png

Space Complexity Graph

Linear System - Positive Definite, Hermitian Matrix - Space.png

Space-Time Tradeoff Improvements

Linear System - Positive Definite, Hermitian Matrix - Pareto Frontier.png