We introduce a detuning-modulated composite pulses control method for robust QIP and implement our solutions in integrated photonic systems. These correct for errors in various parameters, achieving fidelity above the QI threshold 10−4.
We introduce a control method for off-resonant robust quantum information processing suited for quantum integrated photonics. We utilize detunings as control parameters to derive a family of composite pulses for high-fidelity complete population transfer. The presented detuning-modulated composite sequences can be implemented within the decay lifetime of the qubit and correct for control inaccuracies in various parameters including pulse strength, duration, detuning, phase jitter, Stark shift, and unwanted frequency chirps. We implement the proposed robust sequences in an integrated photonics platform to achieve complete light transfer insensitive to fabrication errors.
We present a general method to derive detuning-modualted composite pulses (DMCPs) as N rotations of a canonical two-state quantum system to create accurate and robust pulses that are independent of the initial state of the system. This scheme has minimal pulse overhead, and achieves pulses that are stable against amplitude errors well within the 10−4 threshold that may be suitable for quantum information processing (QIP), within the lifetime of the system. This family of pulses enables to overcome inevitable fabrication errors in silicon photonics, and relax the need for a precise initial state of light coupled into the system to achieve accurate light transfer. Furthermore, we extend universal DMCPs to n-level systems with irreducible SU(2) symmetry to create state transfer that is highly robust to errors in the pulse area from any initial state.
Technologies which harness the laws of quantum mechanics in computational science are pushing toward exponential improvement of information processing tasks. This, as well as innovative functionalities in sensing and imaging, break the fundamental classical limits of computation and measurement. Yet, quantum computation and quantum information processing (QIP) rely on the accuracy of unitary gate operations. In all quantum systems, unitary gates are highly sensitive to deviations from the target control parameters due in part to fabrication errors and environmental changes. Thus, translating quantum information schemes from abstraction to realistic conditions requires robust designs to overcome inherent errors. Moreover, such gates must be universal, meaning that they are independent of the initial state of the qubit system. This adds to the complexity of implementing precise unitary gates that are required for scalable quantum computation. Herein, detuning-modulated composite pulses (DMCPs) are a promising method for achieving accurate and robust state transfer in any qubit hardware platform. We present a DMCP scheme to achieve such universal quantum gates well within the error threshold suitable for QIP. Universal DMCPs are a compelling mechanism for realizing unitary gates in quantum computers under error-prone experimental conditions. These pulses are robust to inaccuracies in pulse strength, duration, resonance offset errors, Stark shifts, etc. within the lifetime of the system.
In this work, we develop a supervised learning model for implementing robust quantum control in composite-pulse systems, where the training parameters can be either phases, detunings, or Rabi frequencies. This model exhibits great resistance to all kinds of systematic errors, including single, multiple, and time-varying errors. We propose a modified gradient descent algorithm for adapting the training of phase parameters, and show that different sampling methods result in different robust performances. In particular, there is a trade-off between high fidelity and robustness for a given number of training parameters, and both can be simultaneously enhanced by increasing the number of training parameters (pulses). For its applications, we demonstrate that the current model can be used for achieving high-fidelity arbitrary superposition states and universal quantum gates in a robust manner. This work provides a highly efficient learning model for fault-tolerant quantum computation by training various physical parameters.