Error Suppression for Arbitrary-Size Black Box Quantum Operations

TL;DR

This paper introduces a scalable error suppression protocol called error filtration (EF) for quantum computations on NISQ devices, which does not require full error correction and works for arbitrary quantum operations.

Contribution

The paper proposes a practical, resource-efficient error suppression method using error filtration that scales independently of the quantum operation size and does not need logical encoding.

Findings

EF provides error suppression when error hierarchy is maintained.

EF is effective for quantum random access memory.

Resources for error suppression are independent of operation size.

Abstract

Efficient suppression of errors without full error correction is crucial for applications with NISQ devices. Error mitigation allows us to suppress errors in extracting expectation values without the need for any error correction code, but its applications are limited to estimating expectation values, and cannot provide us with high-fidelity quantum operations acting on arbitrary quantum states. To address this challenge, we propose to use error filtration (EF) for gate-based quantum computation, as a practical error suppression scheme without resorting to full quantum error correction. The result is a general-purpose error suppression protocol where the resources required to suppress errors scale independently of the size of the quantum operation, and does not require any logical encoding of the operation. The protocol provides error suppression whenever an error hierarchy is respected -- that is, when the ancilliary cSWAP operations are less noisy than the operation to be corrected. We further analyze the application of EF to quantum random access memory, where EF offers hardware-efficient error suppression.