[15p-M_123-1~8] FS.1 Focused Session "AI Electronics"

2:00 PM - 2:15 PM JST(5:00 AM - 5:15 AM UTC)

[15p-M_123-1] Learning Quantum Data Distribution via Chaotic Quantum Diffusion Model

〇Hoan Quoc Tran¹, Koki Chinzei¹, Yasuhiro Endo¹, Hirotaka Oshima¹ (1.Fujitsu Limited)

Keywords:

quantum data distribution,diffusion model,projected ensemble

Generative models for quantum data pose significant challenges but hold immense potential in fields such as chemoinformatics and quantum physics. Quantum denoising diffusion probabilistic models (QuDDPMs) enable efficient learning of quantum data distributions but require high-fidelity random unitary circuits, which demand precise control and exhibit susceptibility to noise on analog hardware. We propose the chaotic quantum diffusion model, a framework that produces projected ensembles generated by chaotic Hamiltonian time evolution, offering a flexible, hardware-compatible diffusion scheme. Requiring only global, time-independent control, our approach reduces implementation overhead across diverse analog quantum platforms while achieving accuracy comparable to that of QuDDPMs. This method enhances trainability and robustness, broadening the applicability of quantum generative models.