Boosting Classification with Quantum-Inspired Augmentations

TL;DR

This paper introduces a quantum-inspired data augmentation technique using small Bloch sphere rotations, which improves image classification accuracy on ImageNet and explores privacy implications, bridging quantum concepts with classical machine learning.

Contribution

The paper presents a novel augmentation method based on SU(2) transformations that enhances classical image classification performance, inspired by quantum gate perturbations.

Findings

Improved Top-1 accuracy by 3% on ImageNet

Enhanced F1 score from 8% to 12% with quantum-inspired augmentation

Stronger unitary transformations can obscure images but do not improve privacy

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.