1
$\begingroup$

I am performing a process tomography over a protocol I suspect to be equivalent to the $CS$ gate.

To compare basic operators I usually compute the Choi matrix of the target gate -- which in this case should be $|CS\rangle\rangle\langle\langle CS|$ -- and check if it is (almost) equivalent to the experimental Choi matrix.

Now, when I try to get the experimental matrix of the circuit I wrote, this has some differences with $|CS\rangle\rangle\langle\langle CS|$.

Is it possible that they are still equivalent, what kind of transformation I am allowed to do to the Choi matrix without changing the channel it is representing?

$\endgroup$

1 Answer 1

2
$\begingroup$

There is a one-to-one relation between Choi states$^1$ and quantum channels (Choi-Jamiołkowski isomorphism). So, you cannot transform the Choi matrix without changing the channel it represents. If the two Choi matrices are different, then they are different quantum channels.


1: Choi states are just normalized Choi matrices. If your map is a valid quantum map, then normalized Choi matrix, i.e. Choi state is a valid quantum state, i.e., the matrix is positive-semidefinite & has unit trace.

$\endgroup$
1
  • 1
    $\begingroup$ What about global phases? Is a Choi matrix global-phase independent? $\endgroup$ Commented Nov 7, 2023 at 17:20

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.