Controlling increasingly complex quantum systems demands faster optimisation techniques; conventional methods often struggle to efficiently shape how qubits evolve over time. A new technique called Quantum Projection Operator-Based Newton Method for Trajectory Optimisation, or Q-PRONTO, improves single-qubit gate design with quicker convergence than existing solvers. The University of Colorado Boulder team has demonstrated Q-PRONTO to improve how qubits are controlled.

This method achieves sharper results than current approaches by utilising a unique approach to finding optimal solutions; it converges on answers more quickly due to its ability to map complex dynamics into simpler functions. The University of Colorado Boulder team developed Q-PRONTO to accelerate qubit control; it’s akin to employing GPS navigation instead of trial and error when charting a course. This new technique speeds up optimisation by transforming complex dynamics into simpler functions, achieving faster convergence than existing methods.

Researchers designed single-qubit gates utilising Q-PRONTO on a fluxonium superconducting qubit, essentially a tiny switch leveraging superconductivity to represent information, and then verified their performance with randomized benchmarking, a standardized test ensuring consistent operation like checking each key on a piano plays the correct note. Modifications were also made to the Quantum Instrumentation Control Kit’s firmware allowing for long sequences of precise pulses. But will this approach scale effectively as quantum systems grow in complexity. Further technical details regarding implementation and results are presented below.

Fluxonium Qubit Gate Errors Reduced Below Fault Tolerance Thresholds via Firmware Optimisation

A University of Colorado Boulder team achieved single-qubit gate errors of just 6.6times10^-4 through a novel optimisation technique, significantly better than typical methods limited to around 1% error. This precision unlocks deeper quantum circuit possibilities by surpassing fault tolerance thresholds demanded by many quantum error correction codes. Achieving these low error rates required firmware modifications for the Quantum Instrumentation Control Kit (QICK), enabling precise extended pulse sequences essential for optimised gates on a fluxonium superconducting qubit, a tiny switch leveraging superconductivity to represent information.

Improved Single-Qubit Fidelity Must Extend To Complex Multi-Qubit Circuits

The new optimisation demonstrably improves single-qubit gate fidelity in fluxonium superconducting qubits, circuits that leverage superconductivity to encode data; however, whether these gains will translate beyond isolated gates remains an open question. Currently, considered a “proof of concept”, optimising more complex multi-qubit operations may present unforeseen challenges. Scaling up from one qubit to many could reveal limitations in Q-PRONTO’s ability to handle increased system interactions and computational demands.

Even acknowledging potential complications when scaling to larger systems, where managing the interplay between numerous quantum bits becomes significantly harder, this represents a step forward for superconducting computing. Validating their approach on a fluxonium circuit (an electronic switch leveraging superconductivity), they measured single-qubit gate accuracy via randomised benchmarking, a standardised test confirming consistent operation; this resulted in 6.6times10^-4 error per gate.

The team will next focus on extending this method to more complex multi-qubit circuits, as achieving such low errors validates Q-PRONTO’s effectiveness at fine-tuning qubit control and offers advantages over existing methods that incrementally improve solutions.

This research demonstrated the successful application of the Quantum Projection Operator-Based Newton Method for Trajectory Optimization (Q-PRONTO) to design high-fidelity gates for a fluxonium superconducting qubit. Using modified firmware with the Quantum Instrumentation Control Kit (QICK), researchers achieved a total error rate of 6.6times10^-4 per single-qubit gate, a level of precision exceeding thresholds required by some quantum error correction codes.

This optimisation method converges more quickly than many current approaches, allowing for finer control of quantum states and potentially enabling longer, more complex circuits. The authors intend to extend this work to multi-qubit systems in future research.

👉 More information
🗞 Optimal Control of a Fluxonium Circuit with the Q-PRONTO Numerical Method
✍️ John Barta, Brian D. Isakov, Pablo Aramburu Sanchez, Jieqiu Shao, Tamás Kalmár, Trevyn F. Q. Larson, Marco M. Nicotra and András Gyenis (University of Colorado Boulder)
🧠 ArXiv: https://arxiv.org/abs/2610.31885

More like this