It is said in some texts, that quantum computer undergoes only 2 types of transofrmations:

1) unitary evolution while computing


2) non-unitary transformation while reading result (output).

What about input?

0) initializing qubits with values

Is it unitary or non-unitary?

  • $\begingroup$ That's the Copenhagen interpretation of quantum mechanics and extends beyond quantum computing. 0) and 2) are identical processes. So far we haven't found counterexamples. Interaction with the environment throws a couple of wrenches into this simplified picture, but we know how to work with those, too. $\endgroup$ – CuriousOne Jan 4 '15 at 16:01
  • 2
    $\begingroup$ Since initialization destroys the previous value, it is not reversible and hence not unitary. The fact that unitary erasure of information is not possible (or the "no-erasure theorem") is the time-reversed counterpart of the famous no-cloning theorem. See nature.com/nature/journal/v404/n6774/abs/404164a0.html and sciencedirect.com/science/article/pii/S0375960103010478 $\endgroup$ – Māris Ozols Jan 4 '15 at 23:47

Initialization is non-unitary (at least if you don't know the state your qubits are in before that). Let's say you want to initialize your qubits in the state $\lvert0\rangle$. First, you need to measure each qubit in some basis (say, $\{\lvert0\rangle,\lvert1\rangle\}$) -- this is the non-unitary part -- and now, if the result was $\lvert1\rangle$, you need to rotate $\lvert1\rangle$ to $\lvert0\rangle$ (which is unitary).


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.