3
$\begingroup$

I am new to the weeds of quantum computing and this question is probably pretty elementary.

Say you run qubit A = (|0⟩ + |1⟩) / √2 and qubit B = |0⟩ through a CNOT gate. What is the state of qubit B afterwards?

Here is how I have tried to reason my way to an answer:

  1. The state of the system after the CNOT gate is (|00⟩ + |11⟩) / √2
  2. The state of qubit A remains (|0⟩ + |1⟩) / √2
  3. Here's where I feel like I'm doing something wrong. My intuition tells me if I have a global state ac|00⟩ + ad|01⟩ + bc|10⟩ + bd|11⟩, I should be able to deconstruct it into the states of the individual qubits to acquire A = a|0⟩ + b|1⟩ and B = c|0⟩ + d|1⟩
  4. Using the above intuition and observations, I have a = 1/√2, b = 1/√2, ad = 0, bc = 0 ac = 1/√2, and bd = 1/√2. I'm looking for c and d.
  5. This system of equations is inconsistent. By the first four equations, it must be the case that c = d = 0. But by the last two, that can't be the case.

Where did my thinking go wrong?

$\endgroup$

1 Answer 1

2
$\begingroup$
  1. This is correct.
  2. This is not correct. Particle A does not have a definite state anymore. Neither does particle B. Previously, you could write the two-particle state as $\frac{|0\rangle+|1\rangle}{\sqrt 2}|0\rangle$, so the state factored and you could talk about "the state of A" and "the state of B". After the CNOT, this is no longer true. You can no longer talk about the "state of A", you can only talk about the state of the SYSTEM. This is because A and B are entangled.

Punchline: There are some two-particle states that cannot be represented as $(a|0\rangle+b|1\rangle)(c|0\rangle+d|1\rangle)$. The state of your A and B particles after applying the CNOT is one of them.

$\endgroup$
6
  • $\begingroup$ Thank you! Is this idea of a particle’s state not being representable as a linear combination of 0 and 1 formalized somehow? Or any pattern behind what separates the different classes of particles? The literature seems to make occasional references to individual particle states after being operated on by a gate (for example, indicating that after a CNOT gate, particle A retains its state and particle B becomes A XOR B.) So I’m just trying to reconcile that with your assertion that the particles have no definite individual state anymore. $\endgroup$
    – jcgrowley
    Commented Mar 14, 2019 at 16:47
  • $\begingroup$ @jcgrowley The issue is not that the particle's state cannot be represented as a combination of 0 and 1, it's that there is no such thing as the particle's state; there's only a state of the overall two-particle system. The two particle system can always be written as a combination of 00, 01, 10, 11, but this can't necessarily be factored into an A- state and a B- state. $\endgroup$ Commented Mar 14, 2019 at 20:18
  • $\begingroup$ @jcgrowley Do a little reading about entanglement in general, and specifically reduced density matrices. Properties of the reduced density matrix will tell you whether or not you can represent a system state as factored into an A- and B- state. I am not sure where exactly you've heard that particle A retains its state after the CNOT gate, but this is emphatically NOT true. It's possible there's something else going on in that statement, so if you have an example I'd be happy to take a look. $\endgroup$ Commented Mar 14, 2019 at 20:19
  • $\begingroup$ I took that from "Quantum Computation and Quantum Information", page 21, figure 1.6, which shows qubits A and B passing through a CNOT gate and becoming A and B ⊕ A. (mmrc.amss.cas.cn/tlb/201702/W020170224608149940643.pdf) I'll definitely read up on reduced density matrices. Thanks for your help. $\endgroup$
    – jcgrowley
    Commented Mar 14, 2019 at 20:25
  • $\begingroup$ @jcgrowley Ah, they're being sloppy! In that diagram (and their formula), they are assuming that the states of $A$ and $B$ are only 0 or 1. In other words, they've ruled out the possibility of sending in the state $(|0\rangle+|1\rangle)|0\rangle$ like you want to do. They are describing the action of CNOT on the basis {00,01,10,11}, and expecting you to extrapolate the rest by linearity. $\endgroup$ Commented Mar 14, 2019 at 20:31

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.