Q-ShiftDP: A Differentially Private Parameter-Shift Rule for Quantum Machine Learning

Hoang M. Ngo, Nhat Hoang-Xuan, Quan Minh Nguyen, Nguyen Hoang Khoi Do, Incheol Shin, My T. Thai
Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, PMLR 300:4447-4455, 2026.

Abstract

Quantum Machine Learning (QML) promises significant computational advantages, but preserving training data privacy remains challenging. Classical approaches like differentially private stochastic gradient descent (DP-SGD) add noise to gradients but fail to exploit the unique properties of quantum gradient estimation. In this work, we introduce the Differentially Private Parameter-Shift Rule (Q-ShiftDP), the first privacy mechanism tailored to QML. By leveraging the inherent boundedness and stochasticity of quantum gradients computed via the parameter-shift rule, Q-ShiftDP enables tighter sensitivity analysis and reduces noise requirements. We combine carefully calibrated Gaussian noise with intrinsic quantum noise to provide formal privacy and utility guarantees, and show that harnessing quantum noise further improves the privacy–utility trade-off. Experiments on benchmark datasets demonstrate that Q-ShiftDP consistently outperforms classical DP methods in QML.

Cite this Paper

BibTeX
@InProceedings{pmlr-v300-ngo26b, title = { Q-ShiftDP: A Differentially Private Parameter-Shift Rule for Quantum Machine Learning }, author = {Ngo, Hoang M. and Hoang-Xuan, Nhat and Nguyen, Quan Minh and Do, Nguyen Hoang Khoi and Shin, Incheol and Thai, My T.}, booktitle = {Proceedings of The 29th International Conference on Artificial Intelligence and Statistics}, pages = {4447--4455}, year = {2026}, editor = {Khan, Emtiyaz and Li, Yingzhen and Solin, Arno and Ramdas, Aaditya}, volume = {300}, series = {Proceedings of Machine Learning Research}, month = {02--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v300/main/assets/ngo26b/ngo26b.pdf}, url = {https://proceedings.mlr.press/v300/ngo26b.html}, abstract = { Quantum Machine Learning (QML) promises significant computational advantages, but preserving training data privacy remains challenging. Classical approaches like differentially private stochastic gradient descent (DP-SGD) add noise to gradients but fail to exploit the unique properties of quantum gradient estimation. In this work, we introduce the Differentially Private Parameter-Shift Rule (Q-ShiftDP), the first privacy mechanism tailored to QML. By leveraging the inherent boundedness and stochasticity of quantum gradients computed via the parameter-shift rule, Q-ShiftDP enables tighter sensitivity analysis and reduces noise requirements. We combine carefully calibrated Gaussian noise with intrinsic quantum noise to provide formal privacy and utility guarantees, and show that harnessing quantum noise further improves the privacy–utility trade-off. Experiments on benchmark datasets demonstrate that Q-ShiftDP consistently outperforms classical DP methods in QML. } }
Endnote
%0 Conference Paper %T Q-ShiftDP: A Differentially Private Parameter-Shift Rule for Quantum Machine Learning %A Hoang M. Ngo %A Nhat Hoang-Xuan %A Quan Minh Nguyen %A Nguyen Hoang Khoi Do %A Incheol Shin %A My T. Thai %B Proceedings of The 29th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2026 %E Emtiyaz Khan %E Yingzhen Li %E Arno Solin %E Aaditya Ramdas %F pmlr-v300-ngo26b %I PMLR %P 4447--4455 %U https://proceedings.mlr.press/v300/ngo26b.html %V 300 %X Quantum Machine Learning (QML) promises significant computational advantages, but preserving training data privacy remains challenging. Classical approaches like differentially private stochastic gradient descent (DP-SGD) add noise to gradients but fail to exploit the unique properties of quantum gradient estimation. In this work, we introduce the Differentially Private Parameter-Shift Rule (Q-ShiftDP), the first privacy mechanism tailored to QML. By leveraging the inherent boundedness and stochasticity of quantum gradients computed via the parameter-shift rule, Q-ShiftDP enables tighter sensitivity analysis and reduces noise requirements. We combine carefully calibrated Gaussian noise with intrinsic quantum noise to provide formal privacy and utility guarantees, and show that harnessing quantum noise further improves the privacy–utility trade-off. Experiments on benchmark datasets demonstrate that Q-ShiftDP consistently outperforms classical DP methods in QML.
APA
Ngo, H.M., Hoang-Xuan, N., Nguyen, Q.M., Do, N.H.K., Shin, I. & Thai, M.T.. (2026). Q-ShiftDP: A Differentially Private Parameter-Shift Rule for Quantum Machine Learning . Proceedings of The 29th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 300:4447-4455 Available from https://proceedings.mlr.press/v300/ngo26b.html.

Related Material