Enhancing Quantum Diffusion Models for Complex Image Generation

TL;DR

This paper introduces a hybrid quantum-classical U-Net model with adaptive observables to improve complex image generation, demonstrating promising results on MNIST despite hardware limitations.

Contribution

It proposes a novel hybrid quantum-classical architecture with adaptive measurements and skip connections for enhanced image generation.

Findings

Successfully generated recognizable MNIST digits across all classes.

Demonstrated potential of hybrid models to mitigate mode collapse.

Highlighted hardware constraints affecting resolution.

Abstract

Quantum generative models offer a novel approach to exploring high-dimensional Hilbert spaces but face significant challenges in scalability and expressibility when applied to multi-modal distributions. In this study, we explore a Hybrid Quantum-Classical U-Net architecture integrated with Adaptive Non-Local Observables (ANO) as a potential solution to these hurdles. By compressing classical data into a dense quantum latent space and utilizing trainable observables, our model aims to extract non-local features that complement classical processing. We also investigate the role of Skip Connections in preserving semantic information during the reverse diffusion process. Experimental results on the full MNIST dataset (digits 0-9) demonstrate that the proposed architecture is capable of generating structurally coherent and recognizable images for all digit classes. While hardware constraints still impose limitations on resolution, our findings suggest that hybrid architectures with adaptive measurements provide a feasible pathway for mitigating mode collapse and enhancing generative capabilities in the NISQ era.