Designing fast quantum gates using optimal control with a reinforcement-learning ansatz

Abstract

Fast quantum gates are crucial not only for the contemporary era of noisy intermediate-scale quantum devices but also for the prospective development of practical fault-tolerant quantum computing. Leakage errors, which arise from data qubits jumping beyond the confines of the computational subspace, are the main challenges in realizing non-adiabatically driven, fast gates. In this work, we propose and illustrate the usefulness of reinforcement learning (RL) to generate fast two-qubit gates in practical multilevel superconducting qubits. In particular, we show that the RL controller offers great effectiveness in finding piecewise constant gate pulse sequences that act on two transmon data qubits coupled by a tunable coupler to generate a controlled-Z (CZ) gate with a gate time of 10 ns and an error rate of $\sim 4\times 10^{-3}$. Using a gradient-based method to solve the same optimization problem often does not achieve high fidelity for such fast gates. However, we show that using the gate pulses discovered by RL as an ansatz for the gradient-based controller can substantially enhance fidelity compared to using RL alone. While for a 10 ns pulse, this improvement is marginal, the combined RL + gradient approach decreases the gate errors below $10^{-4}$ for a gate of length 20 ns.

11 pages, 12 figures

References in corpus (39)

• Quantum Computing in the NISQ era and beyond
• Supplementary information for "Quantum supremacy using a programmable superconducting processor"
• Variational Quantum Algorithms
• Logic gates at the surface code threshold: Superconducting qubits poised for fault-tolerant quantum computing
• Superconducting Qubits: Current State of Play
• Demonstration of Two-Qubit Algorithms with a Superconducting Quantum Processor
• Suppressing quantum errors by scaling a surface code logical qubit
• Real-time quantum error correction beyond break-even
• A tunable coupling scheme for implementing high-fidelity two-qubit gates
• Chopped random-basis quantum optimization
• Repeated Quantum Error Detection in a Surface Code
• Reinforcement Learning in Different Phases of Quantum Control
• Measuring and Suppressing Quantum State Leakage in a Superconducting Qubit
• Realization of high-fidelity CZ and ZZ-free iSWAP gates with a tunable coupler
• Suppression of Qubit Crosstalk in a Tunable Coupling Superconducting Circuit
• Reinforcement Learning with Neural Networks for Quantum Feedback
• Leakage reduction in fast superconducting qubit gates via optimal control
• Tunable Coupling Architecture for Fixed-frequency Transmons
• Fault-tolerant detection of a quantum error
• Artificial Intelligence and Machine Learning for Quantum Technologies
• Learning Quantum Systems
• Quantum Error Correction with the Gottesman-Kitaev-Preskill Code
• High-Fidelity, Frequency-Flexible Two-Qubit Fluxonium Gates with a Transmon Coupler
• Removing leakage-induced correlated errors in superconducting quantum error correction
• Error-transparent operations on a logical qubit protected by quantum error correction
• Overcoming leakage in scalable quantum error correction
• Coherent Transport of Quantum States by Deep Reinforcement Learning
• Measurement Based Feedback Quantum Control With Deep Reinforcement Learning for Double-well Non-linear Potential
• Optimal quantum control of Bose-Einstein condensates in magnetic microtraps: Comparison of GRAPE and Krotov optimization schemes
• Protecting quantum entanglement from leakage and qubit errors via repetitive parity measurements
• Deep Reinforcement Learning Control of Quantum Cartpoles
• Reducing leakage of single-qubit gates for superconducting quantum processors using analytical control pulse envelopes
• Realizing a deep reinforcement learning agent discovering real-time feedback control strategies for a quantum system
• Tunable coupler to fully decouple and maximally localize superconducting qubits
• Optimal generation of Fock states in a weakly nonlinear oscillator
• Errors due to Finite Rise/fall Times of Pulses in Superconducting Charge Qubits
• Robust two-state swap by stimulated Raman adiabatic passage
• Accelerated Magnonic Motional Cooling with Deep Reinforcement Learning
• No-Collapse Accurate Quantum Feedback Control via Conditional State Tomography

Cited by in corpus (1)

• Low-rank optimal control of quantum devices