Numerical Engineering of Robust Adiabatic Operations

TL;DR

This paper introduces a gradient-based optimization protocol that designs robust, fast adiabatic quantum operations tailored to specific experimental imperfections, demonstrated through magnetic resonance and other quantum control applications.

Contribution

A versatile optimization method combining adiabatic control with Hamiltonian engineering, enabling tailored, robust adiabatic operations for various experimental conditions.

Findings

Engineered a fast, robust adiabatic inversion pulse with 99.997% accuracy.

Validated the protocol in nanoscale magnetic resonance experiments.

Provided examples of adiabatic pulses robust to interactions and parameter variations.

Abstract

Adiabatic operations are powerful tools for robust quantum control in numerous fields of physics, chemistry and quantum information science. The inherent robustness due to adiabaticity can, however, be impaired in applications requiring short evolution times. We present a single versatile gradient-based optimization protocol that combines adiabatic control with effective Hamiltonian engineering in order to design adiabatic operations tailored to the specific imperfections and resources of an experimental setup. The practicality of the protocol is demonstrated by engineering a fast, 2.3 Rabi cycle-long adiabatic inversion pulse for magnetic resonance with built-in robustness to Rabi field inhomogeneities and resonance offsets. The performance and robustness of the pulse is validated in a nanoscale force-detected magnetic resonance experiment on a solid-state sample, indicating an ensemble-averaged inversion accuracy of $\sim 99.997\%$. We further showcase the utility of our protocol by providing examples of adiabatic pulses robust to spin-spin interactions, parameter-selective operations and operations connecting arbitrary states, each motivated by experiments.