Quantum Support Vector Regression for Robust Anomaly Detection

2025-05-02

AI Summary

This work addresses semi-supervised anomaly detection in IT security, presenting the first systematic empirical evaluation of Quantum Support Vector Regression (QSVR) on real IBM quantum hardware. Leveraging quantum kernel methods, we conduct experiments across 11 benchmark datasets, incorporating realistic noise modeling—including depolarizing, phase-damping, and bit-flip channels—as well as adversarial example generation to characterize QSVR’s performance limits on NISQ devices. Key findings are: (1) QSVR exhibits robustness under four common noise types and outperforms ideal simulations on two datasets; (2) it is highly sensitive to amplitude-damping noise and calibration errors; (3) it suffers significant adversarial vulnerability, with quantum noise degrading—not enhancing—its adversarial robustness. These results establish the first empirical benchmark for quantum machine learning in security-critical applications and provide critical risk insights for practical deployment.

Technology Category

• Machine Learning: Quantum Machine Learning
• Computer Vision: Adversarial Attacks & Robustness
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Anomaly Detection (AD) is critical in data analysis, particularly within the domain of IT security. In recent years, Machine Learning (ML) algorithms have emerged as a powerful tool for AD in large-scale data. In this study, we explore the potential of quantum ML approaches, specifically quantum kernel methods, for the application to robust AD. We build upon previous work on Quantum Support Vector Regression (QSVR) for semisupervised AD by conducting a comprehensive benchmark on IBM quantum hardware using eleven datasets. Our results demonstrate that QSVR achieves strong classification performance and even outperforms the noiseless simulation on two of these datasets. Moreover, we investigate the influence of - in the NISQ-era inevitable - quantum noise on the performance of the QSVR. Our findings reveal that the model exhibits robustness to depolarizing, phase damping, phase flip, and bit flip noise, while amplitude damping and miscalibration noise prove to be more disruptive. Finally, we explore the domain of Quantum Adversarial Machine Learning and demonstrate that QSVR is highly vulnerable to adversarial attacks and that noise does not improve the adversarial robustness of the model.

Problem

• Exploring quantum ML for robust anomaly detection
• Benchmarking QSVR performance on IBM quantum hardware
• Assessing QSVR vulnerability to adversarial attacks and noise

Innovation

• Quantum Support Vector Regression for anomaly detection
• Benchmarking QSVR on IBM quantum hardware
• Investigating QSVR robustness to quantum noise