ALGORITHMS FOR MATRIX MULTIPLICATION

reportActive / Technical Report | Accession Number: AD0705509 | Open PDF

Abstract:

Strassens and Winograds algorithms for matrix multiplication are investigated and compared with the normal algorithm. Floating - point error bounds are obtained, and it is shown that scaling is essential for numerical accuracy using Winograds method. In practical cases Winograds method appears to be slightly faster than the other two methods, but the gain is, at most, about 20. An attempt to generalize Strassens method is described.

Author(s):
Brent, R. P.
Author Organization(s):
STANFORD UNIV CA DEPT OF COMPUTER SCIENCE
Descriptive Note:
Technical rept.
Supplementary Note:
Sponsored in part by Atomic Energy Commission. DOI: 10.21236/AD0705509
Pagination:
0058

Security Markings

DOCUMENT & CONTEXTUAL SUMMARY

Distribution:
Approved For Public Release, Document Partially Illegible
Distribution Statement:
Approved For Public Release; Distribution Is Unlimited. Document Partially Illegible.

RECORD

Collection: TR
Identifying Numbers
Report Number(s):
STAN-CS-70-157
Contract/Grant Number(s):
N00014-67-A-0112-0029, NSF-GJ-798
Project Number(s):
NR-044-211
Monitor Series:
ONR
 Subject Terms
Joint Capability Areas:
JCA_5_Command and Control; JCA_5.3_Planning; JCA_1_Force Support; JCA_1.3_Human Capital Management
Modernization Areas:
Fully Networked C3
Communities of Interest:
Energy and Power Technologies
Descriptor(s):
*MATRICES(MATHEMATICS), ALGORITHMS, COMPUTER PROGRAMS, ITERATIONS, LEAST SQUARES METHOD, NUMERICAL ANALYSIS
Field(s)/Group(s):
Statistics and Probability, Computer Programming and Software
Keyword(s):
*MATRIX MULTIPLICATION, *MULTIPLICATION
Report Date:
1970 Mar 01