Quantum Denoising Diffusion Models

Kölle, Michael; Stenzel, Gerhard; Stein, Jonas ; Zielinski, Sebastian ; Ommer, Bjorn und Linnhoff-Popien, Claudia (2024): Quantum Denoising Diffusion Models. QSW 2024: IEEE International Conference on Quantum Software, Shenzhen, China, 07. - 13. July 2024. Chang, Rong N.; Chang, Carl K.; Yang, Jingwei; Jin, Zhi; Sheng, Michael; Fan, Jing; Fletcher, Kenneth; He, Qiang; Faro, Ismael; Leymann, Frank; Barzen, Johanna; Puente, Salvador de la; Feld, Sebastian; Wimmer, Manuel; Atukorala, Nimanthi; Wu, Hongyue; Elkouss, David; Garcia-Alonso, Jose und Sarkar, Aritra (eds.) : In: Proceedings 2024 IEEE International Conference on Quantum Software : IEEE QSW 2024, Los Alamitos: IEEE. pp. 88-98

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.

• Item Type: Conference or Workshop Item (Paper)
• Keywords: Measurement ; Image synthesis ; Computational modeling ; Noise reduction ; Computer architecture ; Benchmark testing ; Diffusion models
• Faculties: Mathematics, Computer Science and Statistics > Computer Science
• Subjects: 000 Computer science, information and general works > 004 Data processing computer science
• ISBN: 979-8-3503-6847-5
• Place of Publication: Los Alamitos
• Language: English