Robust Quantum Control for Adiabatic Quantum Computation

TL;DR

This paper introduces an in situ quantum control optimization protocol that enhances adiabatic quantum computation accuracy and robustness without prior spectral gap knowledge, demonstrated on Grover's algorithm and MAX 2-SAT.

Contribution

It develops a novel control optimization method that converges to time-optimal controls and improves fidelity and spectral gap in AQC, robust against system uncertainties.

Findings

Converges to analytically-derived time-optimal controls for Grover's algorithm.

Achieves improved fidelity and spectral gap for MAX 2-SAT.

Demonstrates robustness against control errors.

Abstract

Properly designed control has been shown to be particularly advantageous for improving AQC accuracy and time complexity scaling. Here, an \emph{in situ} quantum control optimization protocol is developed to indirectly optimize state fidelity without knowledge of the instantaneous spectral gap or the computational solution. The protocol is shown to converge to analytically-derived time-optimal controls for Grover's search algorithm. Furthermore, the protocol is utilized to explore optimized control trajectories for the Maximum 2-bit Satisifiability (MAX 2-SAT) problem, where appreciable improvement in fidelity and the minimum spectral gap over a linear schedule is observed. The approach is also shown to be robust against system model uncertainties (unitary control errors). This method is designed to enable robust control optimization on existing quantum annealing hardware and future AQC processors.