Reduction from Matrix Product to Negative Triangle Detection: Difference between revisions

From Algorithm Wiki
Jump to navigation Jump to search
No edit summary
Tags: Manual revert Reverted
No edit summary
Tags: Manual revert Reverted
Line 3: Line 3:
== Description ==  
== Description ==  


$n \times n$ matrices and triangle problem over $R$
Multiplying rectangular matrices
Corollary to generic runtimes


== Implications ==  
== Implications ==  


if: to-time: $O(n^{3}/\log^c n)$ for some constant $c$<br/>then: from-time: $O((\log W) n^{3} / \log^c n)$ where $W$ is maxint of $R$
if: to-time: $T(n)$<br/>then: from-time: $O(mp\cdot T(n^{1/3}) \log W)$ s.t. $mp \leq n^{3}$ and $\sqrt{p} \leq m \leq p^{2}$ and multiplying two matrices of dimensions $m \times n$ and $n \times p$ over $R$ and $W$ is the maxint of $R$


== Year ==  
== Year ==  
Line 18: Line 17:
V. V. Williams, R. R. Williams, Subcubic Equivalences Between Path, Matrix, and Triangle Problems. 2018.
V. V. Williams, R. R. Williams, Subcubic Equivalences Between Path, Matrix, and Triangle Problems. 2018.


https://dl.acm.org/doi/pdf/10.1145/3186893, Corollary 4.1
https://dl.acm.org/doi/pdf/10.1145/3186893, Theorem 4.4

Revision as of 09:58, 10 April 2023

FROM: Matrix Product TO: Negative Triangle Detection

Description

Multiplying rectangular matrices

Implications

if: to-time: $T(n)$
then: from-time: $O(mp\cdot T(n^{1/3}) \log W)$ s.t. $mp \leq n^{3}$ and $\sqrt{p} \leq m \leq p^{2}$ and multiplying two matrices of dimensions $m \times n$ and $n \times p$ over $R$ and $W$ is the maxint of $R$

Year

2018

Reference

V. V. Williams, R. R. Williams, Subcubic Equivalences Between Path, Matrix, and Triangle Problems. 2018.

https://dl.acm.org/doi/pdf/10.1145/3186893, Theorem 4.4