Latent Style-based Quantum GAN for high-quality Image Generation

TL;DR

This paper introduces LaSt-QGAN, a hybrid classical-quantum GAN that generates high-quality images by operating in a latent space, showing promising results on complex datasets and addressing quantum training challenges.

Contribution

The paper presents a novel hybrid classical-quantum GAN architecture that operates in a latent space, enabling high-quality image generation on complex datasets beyond MNIST.

Findings

LaSt-QGAN achieves comparable or better performance than classical GANs on Fashion MNIST and SAT4 datasets.

The approach mitigates barren plateau issues in deep quantum circuits during training.

Empirical and theoretical analysis demonstrate LaSt-QGAN's potential for practical image generation.

Abstract

Quantum generative modeling is among the promising candidates for achieving a practical advantage in data analysis. Nevertheless, one key challenge is to generate large-size images comparable to those generated by their classical counterparts. In this work, we take an initial step in this direction and introduce the Latent Style-based Quantum GAN (LaSt-QGAN), which employs a hybrid classical-quantum approach in training Generative Adversarial Networks (GANs) for arbitrary complex data generation. This novel approach relies on powerful classical auto-encoders to map a high-dimensional original image dataset into a latent representation. The hybrid classical-quantum GAN operates in this latent space to generate an arbitrary number of fake features, which are then passed back to the auto-encoder to reconstruct the original data. Our LaSt-QGAN can be successfully trained on realistic computer vision datasets beyond the standard MNIST, namely Fashion MNIST (fashion products) and SAT4 (Earth Observation images) with 10 qubits, resulting in a comparable performance (and even better in some metrics) with the classical GANs. Moreover, we analyze the barren plateau phenomena within this context of the continuous quantum generative model using a polynomial depth circuit and propose a method to mitigate the detrimental effect during the training of deep-depth networks. Through empirical experiments and theoretical analysis, we demonstrate the potential of LaSt-QGAN for the practical usage in the context of image generation and open the possibility of applying it to a larger dataset in the future.