arXiv cs.AI· Marco Parigi, Stefano Martina, Francesco Aldo Venturelli, Filippo Caruso·· 6 天前AI 评分22
基于真实硬件噪声量子扩散的物理启发生成式 AI 模型
Physics-inspired Generative AI models via real hardware-based noisy quantum diffusion
AI 导读
研究者提出并实现两种物理启发的量子扩散模型协议，用于生成式 AI。第一种利用量子随机游走形式，在前向过程中让量子与经典动力学特定交互，生成 MNIST 图像时 FID 低于纯经典动力学；第二种仅用 4 个 qubit 的 IBM 真实量子硬件，利用其固有噪声生成图像，将量子噪声作为资源而非加以缓解或纠正。
正文 · 原文
Title:Physics-inspired Generative AI models via real hardware-based noisy quantum diffusion
Abstract:Quantum Diffusion Models (QDMs) are an emerging paradigm in Generative AI that aims to use quantum properties to improve the performances of their classical counterparts. However, existing algorithms are not easily scalable due to the limitations of near-term quantum devices. Following our previous work on QDMs, here we propose and implement two physics-inspired protocols. In the first, we use the formalism of quantum stochastic walks, showing that a specific interplay of quantum and classical dynamics in the forward process produces statistically more robust models generating sets of MNIST images with lower Fréchet Inception Distance (FID) than using totally classical dynamics. In the second approach, we realize an algorithm to generate images by exploiting the intrinsic noise of real IBM quantum hardware with only four qubits. Our work could be a starting point to pave the way for new scenarios for large-scale algorithms in quantum Generative AI, where quantum noise is neither mitigated nor corrected, but instead exploited as a useful resource.
| Comments: | 20 pages, 13 figures |
| Subjects: | Quantum Physics (quant-ph); Disordered Systems and Neural Networks (cond-mat.dis-nn); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG) |
| MSC classes: | 81P68, 81P40, 81P47, 68Q12, 68T07, |
| ACM classes: | I.2.6; I.3.3; J.2 |
| Cite as: | arXiv:2505.22193 [quant-ph] |
| (or arXiv:2505.22193v2 [quant-ph] for this version) | |
| https://doi.org/10.48550/arXiv.2505.22193 arXiv-issued DOI via DataCite |
|
| Journal reference: | Parigi, M., Martina, S., Venturelli, F.A. et al. Physics-inspired Generative AI models via real hardware-based noisy quantum diffusion. Quantum Mach. Intell. 8, 110 (2026) |
| Related DOI: | https://doi.org/10.1007/s42484-026-00459-5
DOI(s) linking to related resources |
Submission history
From: Stefano Martina PhD [view email]
[v1]
Wed, 28 May 2025 10:11:48 UTC (4,347 KB)
[v2]
Fri, 9 Oct 2026 12:41:30 UTC (3,242 KB)
来源：arXiv cs.AI · arxiv.org