Quantum Denoising Diffusion Models

TL;DR

This paper proposes quantum diffusion models that leverage quantum machine learning to improve image generation speed and efficiency, outperforming classical models on standard datasets with fewer parameters.

Contribution

Introduction of two quantum diffusion models and a one-step sampling architecture that enhance image generation speed and quality compared to classical diffusion models.

Findings

Quantum models outperform classical counterparts on FID, SSIM, PSNR metrics.

The one-step sampling architecture enables fast image generation.

Models achieve comparable or better results with fewer parameters.

Abstract

In recent years, machine learning models like DALL-E, Craiyon, and Stable Diffusion have gained significant attention for their ability to generate high-resolution images from concise descriptions. Concurrently, quantum computing is showing promising advances, especially with quantum machine learning which capitalizes on quantum mechanics to meet the increasing computational requirements of traditional machine learning algorithms. This paper explores the integration of quantum machine learning and variational quantum circuits to augment the efficacy of diffusion-based image generation models. Specifically, we address two challenges of classical diffusion models: their low sampling speed and the extensive parameter requirements. We introduce two quantum diffusion models and benchmark their capabilities against their classical counterparts using MNIST digits, Fashion MNIST, and CIFAR-10. Our models surpass the classical models with similar parameter counts in terms of performance metrics FID, SSIM, and PSNR. Moreover, we introduce a consistency model unitary single sampling architecture that combines the diffusion procedure into a single step, enabling a fast one-step image generation.