Resilience of bucket-brigade qRAM to spatially and temporally correlated noise

ORAL

Abstract

Quantum random access memory (QRAM) has been presupposed for numerous quantum algorithms and network architectures as an analog to classical RAM. It has been previously shown that the ubiquitous bucket-brigade QRAM is resilient to arbitrary noise channels [PRX Quantum 2, 020311 (2021)], with query infidelities growing only polylogarithmically with memory width when errors are assumed to only occur on individual routers. We consider a generalized, physically-motivated error model consisting of quasi-local correlated errors that affect chains of adjacent routers in the QRAM tree. We show that the QRAM is significantly more resilient to this generalized model than anticipated, and multi-qubit error rates can be exponentially higher than previously assumed while maintaining the favorable error scaling of bucket-brigade QRAM. To conclude, we demonstrate that two-level QRAM, which has been suggested to be similarly noise resilient, can be initialized into arbitrary states and obey notable bounds on the fidelity, indicating that a reset protocol between successive queries may not be necessary.

* AFOSR, ARO (W911NF-18-1-0212), DOE, NSF, NTT Research, and the Packard Foundation.

Presenters

  • Rohan Mehta

    University of Chicago

Authors

  • Rohan Mehta

    University of Chicago

  • Gideon Lee

    University of Chicago

  • Liang Jiang

    University of Chicago