Chalmers TRIO pulses keep superconducting qubit gates at 99.9 percent

Started by Reward Ellie, Today at 07:17 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Topic: Chalmers TRIO pulses keep superconducting qubit gates at 99.9 percent   Views(Read 34 times)
Active members in this topic:
Reward Ellie(1)

Reward Ellie

Researchers at Chalmers University of Technology have shown control pulses that keep qubit gates accurate even when conditions drift. Quantum Zeitgeist's Ivy Delaney reports on the work by Tangyou Huang and colleagues, with partners at CNRS, Université Bourgogne Europe and Aalto University

The team calls the technique TRIO, and it was tested on a two transmon superconducting processor. An X gate reached 99.91 percent fidelity in 90 nanoseconds, while two other gates reached 99.69 and 99.62 percent

The robustness is the real story. With a combined detuning of 4 MHz and a 20 percent amplitude error, TRIO kept fidelity at about 99.5 percent. Standard Gaussian pulses dropped to about 92 percent under the same conditions. The authors describe it as the first experimental demonstration of optimal robust gates

There are caveats. The work is an arXiv preprint and has not been peer reviewed. Pushing gates below 90 nanoseconds also raised leakage to around one in a thousand, and the article's claim of a tenfold improvement does not quite match its own figures

Robust control is valuable because real qubits drift constantly, and recalibrating all the time is expensive. Gates that tolerate small errors could reduce how often machines need tuning. Read alongside the Jülich speed record, it is a good week for superconducting gate design

Is robustness more important than raw fidelity? And how much could pulses like this cut calibration time?