首页 | 本学科首页   官方微博 | 高级检索  
     检索      


Fast linear algebra is stable
Authors:James Demmel  Ioana Dumitriu  Olga Holtz
Institution:(1) Mathematics Department and CS Division, University of California, Berkeley, CA 94720, USA;(2) Mathematics Department, University of Washington, Seattle, WA 98195, USA;(3) Mathematics Department, University of California, Berkeley, CA 94720, USA
Abstract:In Demmel et al. (Numer. Math. 106(2), 199–224, 2007) we showed that a large class of fast recursive matrix multiplication algorithms is stable in a normwise sense, and that in fact if multiplication of n-by-n matrices can be done by any algorithm in O(n ω+η ) operations for any η >  0, then it can be done stably in O(n ω+η ) operations for any η >  0. Here we extend this result to show that essentially all standard linear algebra operations, including LU decomposition, QR decomposition, linear equation solving, matrix inversion, solving least squares problems, (generalized) eigenvalue problems and the singular value decomposition can also be done stably (in a normwise sense) in O(n ω+η ) operations. J. Demmel acknowledges support of NSF under grants CCF-0444486, ACI-00090127, CNS-0325873 and of DOE under grant DE-FC02-01ER25478.
Keywords:Mathematics Subject Classification (2000)" target="_blank">Mathematics Subject Classification (2000)  65F05  65F15  65F25  65F30  65F35  65F40  65G30  65G50  65Y20  68Q25  68W20  68W40  15A52
本文献已被 SpringerLink 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号