Quantum Neural Network Extraction Attack via Split Co-Teaching

Quantum Neural Networks (QNNs), now offered as QNN-as-a-Service (QNNaaS), have become key targets for model extraction attacks. Existing methods use ensemble learning to train substitute QNNs, but our analysis reveals significant limitations in real-world environments, where noise and cost constraints undermine their effectiveness. In this work, we introduce a novel attack, $\mathit{\text{split}}\mathit{\text{co-teaching}}$, which uses label variations to $\mathit{\text{split}}$ queried data by noise sensitivity and employs $\mathit{\text{co-teaching}}$ schemes to enhance extraction accuracy. The experimental results show that our approach outperforms classical extraction attacks by 6.5%$\sim$9.5% and existing QNN extraction methods by 0.1%$\sim$3.7% across various tasks.