1
$\begingroup$

A possibly to ignorant question. Here I read:

For example, if she measures a $|0\rangle$,Bob must measure the same, as $|00\rangle$ is the only state where Alice's qubit is a $|0\rangle$

In the Quantum composer, with the "Bell State ZZ-Measurement" code, q0 and q1 are entagled but, changing the visualization seed (now at 6162), they have distinct measures.

Is this because 'higher correlation' doesn't mean 'identical values' or am I misinterpreting the basic elements?
Many thanks.

Bell State ZZ-Measurement

$\endgroup$
2
  • 2
    $\begingroup$ Correlation between measurement outcomes does not detect entanglement (you might choose to read about Bertlmann's socks!). If you actually want to observe the effect of entanglement, and know it's due to entanglement, you need to do quite a bit more such as measuring a Bell inequality. $\endgroup$
    – DaftWullie
    Commented Jul 8, 2021 at 10:17
  • 2
    $\begingroup$ are you asking the general question written in the title, or about the specific case in the body of the post? $\endgroup$
    – glS
    Commented Jul 8, 2021 at 17:40

1 Answer 1

2
$\begingroup$

Perhaps the simplest way to observe verifiable entanglement between two parties $A$ and $B$ is to have them play the CHSH game. In this game, each player receives an input bit $x$ (resp. $y$) from a referee (also called the verifier), and each player must respond with an output bit $a$ (resp. $b$) to the referee.

Once the game begins, the parties cannot communicate and do not know the inputs given to the other player. The parties win if their outputs satisfy the predicate $a\oplus b=x\cdot y$, or in words, the XOR of their outputs is equal to the AND of their inputs.

One can show that the probability of $A$ and $B$ winning without entanglement between them is at best $3/4$. But if the parties share genuine entanglement, then there is a strategy that wins with probability $\cos^2(\pi/8)\approx 0.85$. Moreover, if they do win with this probability (the game may need to be played over and over to see this), then it can be shown that the parties must be employing genuine quantum correlations (i.e. entanglement).

$\endgroup$
6
  • $\begingroup$ I'm not sure why you need rigidity of the CHSH game to rule out classical strategies. Surely the fact that any non-entangled strategy has a winning probability of at most $3/4$ is sufficient (as long as you have enough statistics). $\endgroup$
    – Rammus
    Commented Jul 8, 2021 at 14:39
  • $\begingroup$ Ah right, rigidity is only needed in the non-ideal case. I will just remove that part of the answer. $\endgroup$
    – Condo
    Commented Jul 8, 2021 at 14:42
  • $\begingroup$ What do you mean by "needed in the non-ideal case"? For detecting entanglement rigidity (self-testing) is never needed as far as I'm aware. $\endgroup$
    – Rammus
    Commented Jul 8, 2021 at 14:51
  • $\begingroup$ Yes, you are right about detection, rigidity is not needed. But to clarify, by "the non-ideal case" I mean when players win with probability at least $\omega-\epsilon$ where $\omega$ is the optimal value and $\epsilon>0$. In this case, the rigidity of CHSH says that their shared state is still "close" to being maximally entangled. In particular, it's $O(\sqrt{\epsilon})$ away from the maximally entangled state in norm. $\endgroup$
    – Condo
    Commented Jul 8, 2021 at 15:13
  • 1
    $\begingroup$ Many thanks @Condo, CHSH game helped me to understand better. I also went trough this video. I understand that probability is at the heart of the question but find it difficult to master it. Recalling the example above. I suppose that this sentence is ill but I do not know how to get out of it: if two qubits are entangled and they are measured in the same basis the two measures will always be identical? $\endgroup$ Commented Jul 11, 2021 at 18:21

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.