Training robust and generalizable quantum models

Abstract

Adversarial robustness and generalization are both crucial properties of reliable machine learning models. In this paper, we study these properties in the context of quantum machine learning based on Lipschitz bounds. We derive parameter-dependent Lipschitz bounds for quantum models with trainable encoding, showing that the norm of the data encoding has a crucial impact on the robustness against data perturbations. Further, we derive a bound on the generalization error which explicitly involves the parameters of the data encoding. Our theoretical findings give rise to a practical strategy for training robust and generalizable quantum models by regularizing the Lipschitz bound in the cost. Further, we show that, for fixed and non-trainable encodings, as those frequently employed in quantum machine learning, the Lipschitz bound cannot be influenced by tuning the parameters. Thus, trainable encodings are crucial for systematically adapting robustness and generalization during training. The practical implications of our theoretical findings are illustrated with numerical results.

References in corpus (31)

• Quantum Computing in the NISQ era and beyond
• Supervised learning with quantum enhanced feature spaces
• Quantum machine learning in feature Hilbert spaces
• Quantum Circuit Learning
• Parameterized quantum circuits as machine learning models
• Circuit-centric quantum classifiers
• The effect of data encoding on the expressive power of variational quantum machine learning models
• Power of data in quantum machine learning
• Challenges and Opportunities in Quantum Machine Learning
• Data re-uploading for a universal quantum classifier
• Robust data encodings for quantum classifiers
• Robust Large Margin Deep Neural Networks
• Quantum machine learning beyond kernel methods
• Generalization in Quantum Machine Learning: a Quantum Information Perspective
• Quantum noise protects quantum classifiers against adversaries
• Quantum Adversarial Machine Learning
• Training robust neural networks using Lipschitz bounds
• Experimental quantum adversarial learning with programmable superconducting qubits
• Machine learning of noise-resilient quantum circuits
• Vulnerability of quantum classification to adversarial perturbations
• Towards quantum enhanced adversarial robustness in machine learning
• Understanding quantum machine learning also requires rethinking generalization
• Shadows of quantum machine learning
• Benchmarking Adversarially Robust Quantum Machine Learning at Scale
• Characterizing the loss landscape of variational quantum circuits
• Generalization despite overfitting in quantum machine learning models
• Robust in Practice: Adversarial Attacks on Quantum Machine Learning
• Optimal Provable Robustness of Quantum Classification via Quantum Hypothesis Testing
• Universal Adversarial Examples and Perturbations for Quantum Classifiers
• Let Quantum Neural Networks Choose Their Own Frequencies
• Robustness of quantum algorithms against coherent control errors

Cited by in corpus (6)

• Adversarial Robustness Guarantees for Quantum Classifiers
• Single-shot quantum machine learning
• Quantum Neural Networks in Practice: A Comparative Study with Classical Models from Standard Data Sets to Industrial Images
• QUACK: Quantum Aligned Centroid Kernel
• Detecting underdetermination in parameterized quantum circuits
• Hybrid quantum recurrent neural network for remaining useful life prediction of turbofan engines