Weighted matrix means and symmetrization procedures (Q1940325): Difference between revisions

The problem of extension of means of two variables to multiple variables has been investigated by several mathematicians; see \textit{J. Lawson} and \textit{Y. Lim} [Math. Ann. 351, No. 2, 267--279 (2011; Zbl 1229.15024)] references cited therein. The author of the present paper in [SIAM J. Matrix Anal. Appl. 32, No. 2, 385--393 (2011; Zbl 1229.15025)] defined means of several matrices based on an iterative procedure, which extends every two-variable matrix mean given by the Kubo-Ando theory (cf. \textit{F. Kubo} and \textit{T. Ando} [Math. Ann. 246, 205--224 (1980; Zbl 0412.47013)]) to several variables. In the present paper, he proves that the procedures of [\textit{T. Ando}, \textit{C.-K. Li} and \textit{R. Mathias}, Linear Algebra Appl. 385, 305--334 (2004; Zbl 1063.47013)] and [\textit{D. A. Bini}, \textit{B. Meini} and \textit{F. Poloni}, Math. Comput. 79, No. 269, 437--452 (2010; Zbl 1194.65065)] converge for any matrix mean. More precisely, it is proved that, for a two-variable function which maps pairs of positive definite matrices to a positive definite matrix and is not greater than the square mean of two positive definite matrices, the Ando-Li-Mathias and Bini-Meini-Poloni procedures converge. Since the Bini-Meini-Poloni procedure requires the existence of a weighted matrix mean, he gives a sufficient definition of a two-variable weighted matrix mean agreeing with the theory of Kubo-Ando.