News

Recent News Archive

Cover of Quantum Robustness in Artificial Intelligence
New Book Chapter

Quantum Robustness in Artificial Intelligence

Our chapter on adversarial robustness in quantum machine learning is now officially available in Springer’s Quantum Robustness in Artificial Intelligence: Principles and Applications.

In collaboration with USTC, we experimentally evaluated the adversarial robustness of 20-qubit quantum neural networks on the Wukong superconducting quantum computer. The study includes training, adversarial attacks, adversarial learning, and robustness certification on both classical and quantum datasets. Results show that QNNs possess intrinsic robustness advantages, and adversarial training significantly enhances their resilience. Our experiments also validate the tightness of theoretical fidelity-based robustness bounds.