首个完全量子化的生成对抗网络,无需经典神经网络即可生成高质量图像。
OrganiQ: Mitigating Classical Resource Bottlenecks of Quantum Generative Adversarial Networks on NISQ-Era Machines
- 纯量子架构设计,摒弃经典神经网络,实现全量子生成对抗。
- 在500次迭代内生成16×16像素图像,峰值信噪比达20.47,质量显著提升。
- 适用于量子计算初学者与希望突破经典瓶颈的量子机器学习研究者。
随着硬件能力的快速进步,量子机器学习已成为研究热点。近期,量子图像生成已取得令人瞩目的成果。然而,以往的量子图像生成方法依赖于经典神经网络,限制了其量子潜力和图像质量。为此,我们提出OrganiQ,首个无需使用经典神经网络即可生成高质量图像的量子生成对抗网络(Quantum GAN)。该方法在500次迭代内实现了16×16像素图像的生成,峰值信噪比(PSNR)达到20.47,显著优于现有方案。实验表明,该架构有效缓解了当前量子设备上的经典资源瓶颈,为未来量子原生图像生成提供了新路径。
原文摘要 · Abstract (English)
Driven by swift progress in hardware capabilities, quantum machine learning has emerged as a research area of interest. Recently, quantum image generation has produced promising results. However, prior quantum image generation techniques rely on classical neural networks, limiting their quantum potential and image quality. To overcome this, we introduce OrganiQ, the first quantum GAN capable of producing high-quality images without using classical neural networks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。