4
$\begingroup$

I am trying to build a circuit for quantum teleportation. On the simulator, everything runs fine and according to expectations, however, I am not able to run the algorithm on the real quantum processor.

ERROR_RUNNING_JOB is returned without any other explanation. I tried to implement two different circuits, the first one with controlled X and Z gates, the second one with gates X and Z are controlled by a value in the classical register. Please find both circuits below.

I appreciate any help.

Thanks.

Circuit with IFs Circuit with "normal" controlled X and Z gates

$\endgroup$
6
$\begingroup$

The issue is that you are applying operations after measurement gates and this is currently not available on the real hardware. I think the hardware also does not support reset operations mid-way through a circuit at the moment.

The best way forward is to keep running this on the simulator or try to find a different way of expressing the circuit such that it avoids these features.

| improve this answer | |
$\endgroup$
  • 3
    $\begingroup$ Also, according to en.wikipedia.org/wiki/Deferred_Measurement_Principle, intermediate measurements are not needed $\endgroup$ – Danylo Y Nov 5 '19 at 13:18
  • $\begingroup$ @DanyloY The deferred measurement principle is a red herring when it comes to communication tasks like teleportation. Yes, you get a circuit with equivalent measurement statistics. But in that new circuit the message was only transmitted at the end of time instead of when you wanted to send it. When you go to use it in real life, e.g. as part of a quantum internet architecture, you will find that deferring all communication indefinitely just doesn't work very well as a communication strategy. $\endgroup$ – Craig Gidney Nov 6 '19 at 23:51
  • $\begingroup$ @CraigGidney Yes, I understand that. But on a (local) quantum processor there is no sense to do intermediate measurements if we are going to use the results immediately. Teleportation is not a plain computational task, it's a communication, so anyway there is no sense to do it just on a single device. $\endgroup$ – Danylo Y Nov 7 '19 at 10:34
1
$\begingroup$

Thanks for help, it works. Here is a changed circuit and results on IBM Q (Vigo processor). Just note that q-bit $0.5|0\rangle + \sqrt {0.75}|1\rangle$ is teleported from q-bit q0 to q2. Angle in Ry gate is $2\pi/3$.

Circuit, Results

| improve this answer | |
$\endgroup$
  • 1
    $\begingroup$ Glad to hear it works! Would you be able to press accept on the answer above so that the question can be marked as answered? Thank you! $\endgroup$ – met927 Nov 6 '19 at 9:05
  • $\begingroup$ I would like to Accept the answer but I do not see any button to do so. $\endgroup$ – Martin Vesely Nov 6 '19 at 11:36
  • 1
    $\begingroup$ This link explains how to accept an answer $\endgroup$ – met927 Nov 6 '19 at 12:46
  • $\begingroup$ Probably, there is something wrong. I do not see check mark under voting. Do you now where can I turn to for help on this? $\endgroup$ – Martin Vesely Nov 7 '19 at 5:55
  • 1
    $\begingroup$ @MartinVesely that's probably because you are not the same user that asked the question (I mean, I get that you are the same person, but the login you used to answer here was different than the one you used to ask the question, as you can see from the different profile images). You need to login with the same account that asked the question. If that was a temporary account, that might not be possible. You can ask a mod to merge the accounts (e.g. by raising a flag on the question and explaining the situation) $\endgroup$ – glS Nov 7 '19 at 11:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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