I remember reading a paper that described how a quantum circuit on a large amount of qubits can be simulated by doing many runs on a smaller quantum computer. They used techniques from tensor networks. I believe it was published in PRL somewhere after 2018, but I haven't been able to find it again. It is not this paper: Computational Speedups Using Small Quantum Devices (Dunjko et al., 2018).

Does anybody know which paper I'm talking about it, or have I simply dreamt up the existence of this paper?

  • $\begingroup$ Is there a restriction on the type of state being simulated? For shapes simulable by a MERA-style circuit (arxiv.org/pdf/quant-ph/0610099.pdf) this sounds feasible. $\endgroup$
    – forky40
    Sep 19, 2019 at 7:17
  • $\begingroup$ The paper concerned universal quantum circuits, and they used some kind of graph theoretic property (where 2-qubit gates form the edges), in order to find right places to 'cut' the circuit. The simulation was not efficient, which is what you would expect if BQP$\neq$ BPP $\endgroup$
    – John
    Sep 20, 2019 at 9:01
  • $\begingroup$ Was is a Divide-and-Conquer approach though? In my group, similar ideas have been mentioned recently $\endgroup$
    – Marsl
    Oct 1, 2019 at 17:38
  • $\begingroup$ @Marsl: I'm not sure what you mean by Divide-and-Conquer. $\endgroup$
    – John
    Oct 3, 2019 at 9:45
  • 1
    $\begingroup$ Is it this one? arxiv.org/pdf/1904.00102.pdf $\endgroup$
    – Marsl
    Oct 15, 2019 at 21:40


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.