I am studying quantum decoding, especially for minimum weight perfect matching. I wonder after finding perfect matching (syndrome graph referred by sparse Blossom), why do we need to select minimum weight? Is there any specific reason for selecting minimum weight?
1 Answer
$\begingroup$
$\endgroup$
5
More weight corresponds to less probability. You're trying to guess the most likely error (max probability, min weight) so you want the min weight matching.
-
$\begingroup$ Would you please explain more ? Why more weigth corresponds to less probability? $\endgroup$– 김동민Commented Feb 14 at 0:29
-
1$\begingroup$ @김동민 because w = -log(p / (1 - p)). When p is near 0, w increases as p decreases. $\endgroup$ Commented Feb 14 at 0:40
-
$\begingroup$ Where does the equation come from? paper? thesis? $\endgroup$– 김동민Commented Feb 14 at 0:42
-
-
1