Re: Matrix Minimax, sorta
- From: israel@xxxxxxxxxxx (Robert Israel)
- Date: 20 Oct 2005 07:15:03 GMT
In article <dj572n$9bs$2@xxxxxxxxxxxxxxxxxxxxxxxxxx>,
Hauke Reddmann <fc3a501@xxxxxxxxxxxxxx> wrote:
>I have the equation aA+bB+cC+...=min.,
>where a,b,c... are scalars, A,B,C are matrices,
>a=b=c=...=0 is ruled out and "min." isn't
>Euclidean norm but the rank.
>
>For special cases (e.g. two 2x2 matrices - just
>fix a/b so that the determinant vanishes) this
>is easy, but can you give a general approach?
The rank is the size of the largest nonsingular
square submatrix. For the rank to be <= r, all
the (r+1) x (r+1) submatrices must have
determinant 0. Those determinants are polynomials
in a,b,c,... So it's essentially reduced to
solving a (maybe rather large) set of polynomials.
Groebner basis techniques may help.
Robert Israel israel@xxxxxxxxxxx
Department of Mathematics http://www.math.ubc.ca/~israel
University of British Columbia Vancouver, BC, Canada
.
- References:
- Matrix Minimax, sorta
- From: Hauke Reddmann
- Matrix Minimax, sorta
- Prev by Date: Re: relation from A to B. question about prove
- Next by Date: Re: getting rid of powers in factoring
- Previous by thread: Re: Matrix Minimax, sorta
- Next by thread: Correction for: Interpolating function with two important characteristics
- Index(es):
Relevant Pages
|