I have to ptove that $O(n,\mathbb{R})$ is a strong deformation retract of $GL(n,\mathbb{R})$. In order to do this I must check the continuity of the map $GL(n,\mathbb{R})\longrightarrow O(n,\mathbb{R})$ defined by $A=O\cdot S\longmapsto O$ ,where $O\in O(n,\mathbb{R})$ and $S$ is real symmetric and positively defined. How I can check the continuity?
Continuity of a map between matrices
2
$\begingroup$
matrices
algebraic-topology
lie-groups
homotopy-theory
1 Answers
2
The matrix $S$ can be explicitly obtained from $A$ as $\sqrt{A^TA}$, where the radical sign means the unique positive-definite square root. So it suffices to show that the map taking a positive-definite matrix to its positive-definite square root is continuous. There are several ways you can do this. For instance, if $B$ is a positive-definite matrix whose spectrum is contained in $(r,R)$ for $0 (Note that a more elementary deformation retraction from $GL(n,\mathbb{R})$ to $O(n,\mathbb{R})$ can be obtained by applying Gram-Schmidt orthonormalization to the columns of $A$. Each step of Gram-Schmidt is easily seen to be continuous and can be interpolated by a homotopy.)
-
0Do you think that there is a moore direct proof? – 2017-02-12
-
0Possibly, but I don't know one off the top of my head. – 2017-02-12