2
$\begingroup$

For $p \in [1, \infty)$, the $p$-Schatten norm of a matrix $A$ is defined as $\|A\|_p = Tr(\left|A\right|^p)^{\frac{1}{p}}$, where $\left|A\right| = (A^* A)^{\frac{1}{2}}$. Is there any function to calculate those norms?

$\endgroup$

1 Answer 1

2
$\begingroup$
Total[SingularValueList[A]^p]^(1/p)

or

Power[Tr[MatrixPower[A\[Transpose] . A, p/2]], 1/p]
$\endgroup$
4
  • $\begingroup$ The first one is a bit faster. $\endgroup$
    – Roman
    Commented Feb 18, 2022 at 20:53
  • $\begingroup$ Thank you very much! $\endgroup$
    – Seven9
    Commented Feb 18, 2022 at 21:06
  • $\begingroup$ Why not use Norm[...,p] on the SingularValueList? $\endgroup$
    – The Vee
    Commented Aug 1, 2022 at 8:31
  • 1
    $\begingroup$ Bevause Norm is slower. $\endgroup$ Commented Aug 1, 2022 at 8:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.