arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2607.03337quant-ph
收藏

对量子神经网络的端到端多阶段杀伤链攻击:在囚禁离子硬件上的演示

An End-to-End Multi-Stage Kill-Chain Attack on Quantum Neural Networks: Demonstration on Trapped-Ion Hardware

d-fine GmbH(d-fine 有限公司);Fraunhofer Institute for Applied and Integrated Security (AISEC)(弗劳恩霍夫应用和综合安全研究所);Alpine Quantum Technologies (AQT) GmbH(阿尔卑斯量子技术 有限公司);Federal Office for Information Security (BSI)(联邦信息安全局)

Cedric Brügmann, Daniel Herr, Daniel Ohl de Mello, Pascal Debus, Maximilian Wendlinger, Kilian Tscharke, Juris Ulmanis, Alexander Erhard, Arthur Schmidt, Fabian Petsch

AI总结:

研究针对囚禁离子量子计算机上运行的量子神经网络模型的端到端多阶段攻击,通过侧信道侦察、串扰表征、对抗样本生成及物理串扰攻击实现,覆盖离子阱完整攻击链并讨论对量子即服务提供商及硬件缓解措施的影响。

摘要:

我们展示了对在囚禁离子量子计算机上执行的量子神经网络(QNN)模型的端到端、多阶段攻击。我们的链结合了侧信道侦察、串扰表征、对抗样本生成以及在设备上实现对抗性扰动的物理串扰攻击。我们涵盖了离子阱上的完整攻击链,并在附录中报告了相应的超导硬件实验。我们讨论了对量子即服务提供商和硬件缓解措施的影响。

英文摘要:

We demonstrate an end-to-end, multi-stage attack against a quantum neural network (QNN) model that is executed on a trapped-ion quantum computer. Our chain combines side-channel reconnaissance, crosstalk characterization, adversarial example generation, and a physical crosstalk attack that realizes the adversarial perturbation on the device. We cover the full attack chain on ion traps and report the corresponding superconducting-hardware experiments in the appendix. We discuss implications for QaaS providers and hardware mitigations.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑