소재 AI 공유함
Menu
← AI in materials

AI trends · written by AI

Fast and robust superconducting quantum gate pulse design with Neural ODE

AI editorPublished 2026.09.29

A Neural ODE based on neural networks directly generated control pulses and designed and experimentally implemented a π/2 gate in a superconducting transmon circuit. The reported performance was about 99.95% gate fidelity and detuning robust performance at the ±20 MHz level. [1][2]

1/ AI method: Neural ODE based neural network
2/ Applied task: design of quantum gate control pulses for a superconducting transmon circuit
3/ Reported results: implementation of a short, detuning-robust π/2 pulse, about 99.95% fidelity and 99.9% or higher parity measurement efficiency
4/ This study in Physical Review Applied used Neural ODE to design fast and robust quantum gate control pulses and implement them in an actual circuit.
5/ The target was a π/2 pulse for a superconducting transmon circuit, and the neural network was used to optimize by directly generating the control waveform.
6/ According to the report, the implemented gate showed about 99.95% fidelity, the parity measurement efficiency was 99.9% or higher, and robust performance was maintained over a detuning range of about ±20 MHz.
7/ This case can be seen as an application in which AI, although not designing materials themselves, solved a real control problem on a materials and device experimental platform.
8/ However, because this result was tailored to a specific superconducting quantum device and control condition, the same performance cannot be expected simply by transferring it to other hardware or larger circuits. [1][2]

How to apply this

You can try designing short, noise-resistant pulses for a desired operation with a neural-network-based control model.

  1. Prepare a simulation environment that reflects the device dynamics and control constraints.
  2. Train a control pulse that produces the target operation with a continuous-time neural network such as Neural ODE.
  3. Stress-test candidate pulses under detuning and noise conditions to evaluate robustness.
  4. Apply them to the experimental device and independently verify fidelity and measurement efficiency.
Data needed
The required data include the Hamiltonian of the target device, the allowable control waveform range, and training simulation or measured response data.
Checking the result
Test the designed pulse in independent simulation and on the actual device, and separately verify the fidelity and noise sensitivity of the target gate.
What can go wrong
If the training data are insufficient or the device model is inaccurate, performance may look good in simulation but drop sharply on the real device.

이 가이드는 원문의 방법을 바탕으로 AI가 제안한 적용 방향입니다. 원문에 적힌 내용이 아니며 같은 결과가 재현된다는 보장도 아닙니다.

Related concepts and practice

이 이슈를 읽는 데 필요한 개념입니다. 용어마다 Colab에서 바로 돌려볼 수 있는 실습 노트북이 있습니다.

Sources

  1. Physical Review Applied - Accepted Paper: Neural-network-based design and implementation of fast and robust quantum gates
  2. Neural-network-based design and implementation of fast and robust quantum gates

웹 자료를 바탕으로 AI가 자동 요약한 글입니다. 연구의 세부 조건과 결과는 연결된 원문을 확인해주세요.