Superior resilience to poisoning and amenability to unlearning in quantum machine learning

2025-08-04

AI Summary

This work addresses the reliability of AI models trained on noisy and poisoned data by systematically comparing the robustness and machine unlearning capabilities of classical versus quantum neural networks. We inject label noise into both classical and quantum datasets and perform approximate unlearning operations. Our experiments reveal that quantum neural networks exhibit a phase-transition-like response to noise, achieving significantly superior generalization performance in high-noise regimes compared to classical counterparts. We further introduce the first quantum machine unlearning framework, enabling reversible and computationally efficient data deletion—overcoming the irreversibility and prohibitive computational overhead inherent in classical unlearning methods. Empirical results demonstrate that quantum models simultaneously attain enhanced noise resilience and dynamic adaptability. This work establishes a novel paradigm for building trustworthy, auditable, quantum-enhanced learning systems.

Technology Category

• Machine Learning: Quantum Machine Learning
• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Learning to Search

Application Category

• Security and Privacy: Security and privacy of machine learning and AI applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

The reliability of artificial intelligence hinges on the integrity of its training data, a foundation often compromised by noise and corruption. Here, through a comparative study of classical and quantum neural networks on both classical and quantum data, we reveal a fundamental difference in their response to data corruption. We find that classical models exhibit brittle memorization, leading to a failure in generalization. In contrast, quantum models demonstrate remarkable resilience, which is underscored by a phase transition-like response to increasing label noise, revealing a critical point beyond which the model's performance changes qualitatively. We further establish and investigate the field of quantum machine unlearning, the process of efficiently forcing a trained model to forget corrupting influences. We show that the brittle nature of the classical model forms rigid, stubborn memories of erroneous data, making efficient unlearning challenging, while the quantum model is significantly more amenable to efficient forgetting with approximate unlearning methods. Our findings establish that quantum machine learning can possess a dual advantage of intrinsic resilience and efficient adaptability, providing a promising paradigm for the trustworthy and robust artificial intelligence of the future.

Problem

• Comparing resilience to data corruption in classical vs quantum neural networks
• Exploring quantum machine unlearning for efficient forgetting of corrupt data
• Demonstrating quantum models' superior adaptability and robustness in AI

Innovation

• Quantum models show resilience to data corruption
• Quantum machine unlearning enables efficient forgetting
• Phase transition-like response to label noise