摘要: Introduction In numerical linear algebra one meets condition numbers [[AII []A-11[ and similar quantities such as (max [a,i[)[[A-1H [[Ai[[ [[A-l[], where A = (a,i) i is the /'-th column of A. The norms are very diverse. problem then to determine a rowand/or column-scaling which minimizes quanti ty under consideration. I t purpose this paper specify class for those scalings can be given explicitly. results will extensions some in [2]. They also hold non-square matrices. All proofs completely elementary. Also, cases minimizing scaling cannot explicitly, it said how far at most certain consideration may away from its minimum.
springer.com
acm.org
uu.se 
sci-hub.se 参考文章(6)
F. L. Bauer, Optimally scaled matrices Numerische Mathematik. ,vol. 5, pp. 73- 87 ,(1963) , 10.1007/BF01385880
F. L. Bauer, J. Stoer, C. Witzgall, Absolute and monotonic norms Numerische Mathematik. ,vol. 3, pp. 257- 264 ,(1961) , 10.1007/BF01386026
George E Forsythe, Ernst G Straus, None, On best conditioned matrices Proceedings of the American Mathematical Society. ,vol. 6, pp. 340- 345 ,(1955) , 10.1090/S0002-9939-1955-0069585-4
George E. Forsythe, Wolfgang R. Wasow, William Nachbar, Finite-Difference Methods for Partial Differential Equations ,(2004)
Alexander Ostrowski, Über Normen von Matrizen. Mathematische Zeitschrift. ,vol. 63, pp. 2- 18 ,(1955) , 10.1007/BF01187920
George E. Forsythe, Wolfgang R. Wasow, Finite-difference methods for partial differential equations Applied Mathematics Series. ,(1960)