Image Classification with Rotation-Invariant Variational Quantum Circuits

TL;DR

This paper introduces a rotation-invariant variational quantum classifier architecture for image classification, demonstrating improved performance through geometric bias and extending it with classical convolution for larger images on NISQ devices.

Contribution

It proposes a novel equivariant quantum classifier architecture with rotational invariance and extends it with classical convolution to handle larger images.

Findings

Geometric bias enhances quantum model performance.

The equivariant architecture outperforms non-equivariant benchmarks.

Classical convolution extends quantum models to larger images.

Abstract

Variational quantum algorithms are gaining attention as an early application of Noisy Intermediate-Scale Quantum (NISQ) devices. One of the main problems of variational methods lies in the phenomenon of Barren Plateaus, present in the optimization of variational parameters. Adding geometric inductive bias to the quantum models has been proposed as a potential solution to mitigate this problem, leading to a new field called Geometric Quantum Machine Learning. In this work, an equivariant architecture for variational quantum classifiers is introduced to create a label-invariant model for image classification with $C_4$ rotational label symmetry. The equivariant circuit is benchmarked against two different architectures, and it is experimentally observed that the geometric approach boosts the model's performance. Finally, a classical equivariant convolution operation is proposed to extend the quantum model for the processing of larger images, employing the resources available in NISQ devices.