Quantum enhancement in generative models

Francisco Fernandez Alcazar, Alejandro Perdomo‐Ortiz

Jincheng Zhang

This paper explores the potential of quantum-enhanced generative modeling through the development of novel variational circuit-based models. We propose a new class of generative models that exploit quantum effects to improve sampling efficiency and feature extraction capabilities, moving beyond the limitations of classical generative techniques. The core mechanism centers on employing variational circuits to generate complex data, pushing the boundaries of what's currently achievable in generative modeling. The integration of quantum chemistry principles into variational circuit design allows for a pathway towards significantly enhanced data quality and representation. This work presents a theoretical framework and initial explorations demonstrating the potential of this approach, highlighting the advantages of quantum-enhanced generative modeling.

Madhusudan Singh Irish Singh, Bharat S. Rawal

Nora M. Bauer, Kübra Yeter Aydeniz, Max F. Masnick

Generative learning models in medical research are crucial in developing training data for deep learning models and advancing diagnostic tools, but the problem of high quality, diverse images is an open topic of research. Quantum-enhanced generative models have been proposed and tested in the literature but have been restricted to small problems below the scale of industry relevance. In this paper, we propose quantum-enhanced diffusion and variational autoencoder (VAE) models and test them on the Fundus retinal image generation task. In our numerical experiments, the images generated using quantum-enhanced models are of higher quality and match more closely in features to the real image distribution compared to the ones generated using classical diffusion models, even when the classical diffusion models are larger than the quantum model. Additionally, we perform noisy testing and experimental runs on hardware to confirm the numerical experiments, finding that quantum-enhanced diffusion model can sometimes produce higher quality images, both in terms of diversity and fidelity, when tested with quantum noise.

Kusal M. Abeywickrama, Srinjoy Ganguly, Luis Gerardo Ayala Bertel +1 more

Juan Carrasquilla, Giacomo Torlai, Roger G. Melko +1 more