Скачать или смотреть Software Mitigation of Crosstalk on Noisy Intermediate-Scale Quantum Computers

Full paper: https://mrmgroup.cs.princeton.edu/pap...

Prakash Murali, David C. McKay, Margaret Martonosi, Ali Javadi Abhari "Software Mitigation of Crosstalk on Noisy Intermediate-Scale Quantum Computers", International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), 2020.

Code: https://github.com/Qiskit/qiskit-terr...

https://github.com/Qiskit/qiskit-terr...

Abstract: In recent years, QC hardware has progressed to the point where Noisy Intermediate-Scale Quantum systems (NISQ) are available for real-system application studies. To perform useful computation on these devices, it is crucial to develop hardware/software approaches which extract the most reliable executions from noisy hardware. Crosstalk is a major source of noise in NISQ systems and is a fundamental challenge for hardware design. When multiple instructions are executed in parallel, crosstalk among the instructions can corrupt the quantum state and lead to incorrect program execution. Our goal is to mitigate the application impact of crosstalk noise through software techniques.

Crosstalk mitigation requires an accurate characterization of the hardware crosstalk noise, and also, intelligent instruction scheduling to serialize operations affected by crosstalk. Since crosstalk characterization is computationally expensive, we develop optimizations which reduce the characterization overhead. On 3 20-qubit IBMQ systems, we demonstrate two orders of magnitude reduction in characterization time (compute time on the QC device) compared to all-pairs crosstalk measurements. Enabled by our characterization, we develop an instruction scheduler that balances the need to mitigate crosstalk as well as decoherence (qubit loss of information when instructions are delayed). Our scheduler serializes high crosstalk instructions to minimize crosstalk, but, simultaneously minimizes circuit length. On real-system runs on 3 IBMQ systems, our scheduler improves the error rate of parallel SWAP circuits by up to 5.6x, compared to the IBM instruction scheduler. Since SWAP operations are a fundamental communication primitive in superconducting QC systems, the benefit offered by crosstalk-adaptivity is an important advance on the path to reliable QC execution.

In a broader picture, the difficulty of mitigating crosstalk has recently driven QC vendors to move towards sparser qubit connectivity or disabling nearby operations entirely in hardware, which can be detrimental to performance. Our work makes the case for software mitigation of crosstalk errors.