Implementing Large Quantum Boltzmann Machines as Generative AI Models for Dataset Balancing

2025-02-05

AI Summary

To address severe class imbalance among attack categories in intrusion detection systems (IDS), this paper proposes a generative data augmentation method based on a large-scale quantum restricted Boltzmann machine (QRBM) implemented on D-Wave’s Pegasus quantum annealing hardware. We achieve the first successful custom embedding of a 120×120 QRBM on the Pegasus architecture—overcoming limitations of default embedding tools—and generate over 1.6 million high-fidelity synthetic attack samples within milliseconds, yielding a balanced 4.2-million-sample dataset. Compared to classical oversampling methods such as SMOTE, downstream classifiers trained on our augmented data exhibit statistically significant improvements in detection rate, precision, recall, and F1-score. To date, this work represents the largest-scale QRBM implementation and the first end-to-end deployment and empirical validation of a generative QRBM model on Pegasus hardware. It establishes a novel paradigm for leveraging quantum machine learning to enhance cybersecurity capabilities.

Technology Category

• Machine Learning: Quantum Machine Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

This study explores the implementation of large Quantum Restricted Boltzmann Machines (QRBMs), a key advancement in Quantum Machine Learning (QML), as generative models on D-Wave's Pegasus quantum hardware to address dataset imbalance in Intrusion Detection Systems (IDS). By leveraging Pegasus's enhanced connectivity and computational capabilities, a QRBM with 120 visible and 120 hidden units was successfully embedded, surpassing the limitations of default embedding tools. The QRBM synthesized over 1.6 million attack samples, achieving a balanced dataset of over 4.2 million records. Comparative evaluations with traditional balancing methods, such as SMOTE and RandomOversampler, revealed that QRBMs produced higher-quality synthetic samples, significantly improving detection rates, precision, recall, and F1 score across diverse classifiers. The study underscores the scalability and efficiency of QRBMs, completing balancing tasks in milliseconds. These findings highlight the transformative potential of QML and QRBMs as next-generation tools in data preprocessing, offering robust solutions for complex computational challenges in modern information systems.

Problem

• Quantum Boltzmann Machines for dataset imbalance
• Improving Intrusion Detection Systems with quantum models
• Generating synthetic samples to enhance classifier performance

Innovation

• Quantum Restricted Boltzmann Machines
• Pegasus quantum hardware
• Dataset balancing efficiency