Table of Contents
Fetching ...

Implementation and verification of coherent error suppression using randomized compiling for Grover's algorithm on a trapped-ion device

Masatoshi Ishii, Hammam Qassim, Tomochika Kurita, Joseph Emerson, Kazunori Maruyama, Hirotaka Oshima, Shintaro Sato

Abstract

In near-term quantum computations that do not employ error correction, noise can proliferate rapidly, corrupting the quantum state and making results unreliable. These errors originate from both decoherence and control imprecision. The latter can manifest as coherent noise that is especially detrimental. Here, we study the impact of coherent errors and their mitigation under standard error-reduction techniques, both theoretically and experimentally on a trapped-ion quantum computer. As a representative case study, we implement a range of Grover's algorithm circuits containing up to 10 qubits and 26 two-qubit gates. We demonstrate the effectiveness of randomized compiling (RC) and algorithm error detection (ED), where the latter is realized via post-selection on ancillary qubits that ideally return to the ground state at the end of each circuit. Our results highlight a synergetic effect: combining RC and ED yields the largest reductions in errors, indicating that these methods can work together to extend the capabilities of near-term quantum devices for moderately deep circuits.

Implementation and verification of coherent error suppression using randomized compiling for Grover's algorithm on a trapped-ion device

Abstract

In near-term quantum computations that do not employ error correction, noise can proliferate rapidly, corrupting the quantum state and making results unreliable. These errors originate from both decoherence and control imprecision. The latter can manifest as coherent noise that is especially detrimental. Here, we study the impact of coherent errors and their mitigation under standard error-reduction techniques, both theoretically and experimentally on a trapped-ion quantum computer. As a representative case study, we implement a range of Grover's algorithm circuits containing up to 10 qubits and 26 two-qubit gates. We demonstrate the effectiveness of randomized compiling (RC) and algorithm error detection (ED), where the latter is realized via post-selection on ancillary qubits that ideally return to the ground state at the end of each circuit. Our results highlight a synergetic effect: combining RC and ED yields the largest reductions in errors, indicating that these methods can work together to extend the capabilities of near-term quantum devices for moderately deep circuits.

Paper Structure

This paper contains 19 sections, 9 equations, 12 figures, 1 table.

Figures (12)

  • Figure 1: Simulation results of TVD for the number of marked solutions without any error suppression. (The TVD for 1000 randomly selected different marked bitstrings oracle with the same number of marked solutions are plotted. Comparison under two noise models: Over-rotational (OR) noise only, and a combination of OR and decoherence noises.) The TVD due to over-rotation noise are observed to exhibit greater dependence on the number of marked solutions and the marked bitstring oracle.
  • Figure 2: Simulation results on marked bitstring oracle dependency in TVD under only over-rotation noise, with 16 marked solution. (Comparison between a randomly selected marked bitstring oracle and an oracle circuit composed of a different number of $CZ$ gates.) Numerical data shows that oracles implementation with three $CZ$ gates have a large distribution of TVD values and almost the same variance as a randomly selected marked bitstring oracle.
  • Figure 3: Experimental results for a circuit implementing three $CZ$ gate oracles circuits using a trapped-ion device, with 16 marked solutions and randomly selected different marked bitstring oracles.
  • Figure 4: Simulation results for TVD variations based on the number of marked solutions for randomly selected marked bitstring oracles for cases without error suppression (ES), with randomized compiling (RC), with error detection (ED), and with both RC and ED.
  • Figure 5: Analysis results of quantum error suppression mechanisms in simulation. (This figure presents an alternative analysis of the same data as Figure \ref{['fig:NumofSol_tvd']} highlights the effect of error suppression at the level of individual different marked bitstring oracles, which cannot be shown in a violin plot.)
  • ...and 7 more figures