Boosting Classification with Quantum-Inspired Augmentations

2025-06-27

AI Summary

Conventional data augmentation methods lack spatial interpretability, and existing quantum-inspired approaches often require quantum hardware or trainable modules, limiting practical applicability in classical image classification. Method: This work proposes a quantum-inspired data augmentation technique—small-angle random Bloch sphere rotations—applying SU(2) unitary transformations directly to classically encoded complex-valued pixel representations, without quantum hardware or learnable parameters. Contribution/Results: On ImageNet, the method improves Top-1 accuracy by 3.0%, Top-5 by 2.5%, and F₁ score from 8% to 12%, demonstrating that infinitesimal quantum perturbations effectively enhance classical models. Further analysis reveals that while such strong unitary transformations exhibit potential for privacy-preserving computation, they fail to satisfy differential privacy guarantees—exposing an inherent utility–privacy trade-off. Crucially, this is the first systematic integration of a non-learnable, geometrically explicit quantum state rotation mechanism into classical data augmentation.

Technology Category

• Machine Learning: Quantum Machine Learning
• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Non-convex Optimization

Application Category

• Security and Privacy: Data transparency and provenance
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Understanding the impact of small quantum gate perturbations, which are common in quantum digital devices but absent in classical computers, is crucial for identifying potential advantages in quantum machine learning. While these perturbations are typically seen as detrimental to quantum computation, they can actually enhance performance by serving as a natural source of data augmentation. Additionally, they can often be efficiently simulated on classical hardware, enabling quantum-inspired approaches to improve classical machine learning methods. In this paper, we investigate random Bloch sphere rotations, which are fundamental SU(2) transformations, as a simple yet effective quantum-inspired data augmentation technique. Unlike conventional augmentations such as flipping, rotating, or cropping, quantum transformations lack intuitive spatial interpretations, making their application to tasks like image classification less straightforward. While common quantum augmentation methods rely on applying quantum models or trainable quanvolutional layers to classical datasets, we focus on the direct application of small-angle Bloch rotations and their effect on classical data. Using the large-scale ImageNet dataset, we demonstrate that our quantum-inspired augmentation method improves image classification performance, increasing Top-1 accuracy by 3%, Top-5 accuracy by 2.5%, and the F$_1$ score from 8% to 12% compared to standard classical augmentation methods. Finally, we examine the use of stronger unitary augmentations. Although these transformations preserve information in principle, they result in visually unrecognizable images with potential applications for privacy computations. However, we show that our augmentation approach and simple SU(2) transformations do not enhance differential privacy and discuss the implications of this limitation.

Problem

• Investigates quantum-inspired data augmentation for classical machine learning
• Explores small-angle Bloch rotations' impact on image classification performance
• Examines unitary augmentations' potential for privacy-preserving computations

Innovation

• Quantum-inspired data augmentation boosts classical ML
• Random Bloch sphere rotations enhance image classification
• Simulated small-angle SU(2) transformations improve accuracy