扩散模型：方法与应用综述

Diffusion Models: A Comprehensive Survey of Methods and Applications

作者： Ling Yang, Zhilong Zhang, Shenda Hong, Runsheng Xu, Yue Zhao, Yingxia Shao, Wentao Zhang, Ming-Hsuan Yang, Bin Cui

https://paperswithcode.com/paper/diffusion-models-a-comprehensive-survey-of

https://arxiv.org/abs/2209.00796

扩散模型是一类深层生成模型，在具有坚实理论基础的各种任务上显示了令人印象深刻的结果。尽管证明比最先进的方法更成功，扩散模型通常需要昂贵的采样程序和次优似然估计。已经作出了重大努力来改进扩散模型在各个方面的性能。在本文中，我们对扩散模型的现有变体进行了全面回顾。具体而言，我们提供了扩散模型的分类，并将其分为三种类型：采样加速增强、似然最大化增强和数据泛化增强。我们还介绍了其他生成模型（即变分自动编码器、生成对抗网络、归一化流、自回归模型和基于能量的模型），并讨论了扩散模型和这些生成模型之间的联系。然后，我们回顾了扩散模型的应用，包括计算机视觉、自然语言处理、波形信号处理、多模态建模、分子图生成、时间序列建模和对抗性净化。此外，我们提出了有关生成模型开发的新观点。github：

Diffusion models are a class of deep generative models that have shown impressive results on various tasks with a solid theoretical foundation. Despite demonstrated success than state-of-the-art approaches, diffusion models often entail costly sampling procedures and sub-optimal likelihood estimation. Significant efforts have been made to improve the performance of diffusion models in various aspects. In this article, we present a comprehensive review of existing variants of diffusion models. Specifically, we provide the taxonomy of diffusion models and categorize them into three types: sampling-acceleration enhancement, likelihood-maximization enhancement, and data-generalization enhancement. We also introduce the other generative models (i.e., variational autoencoders, generative adversarial networks, normalizing flow, autoregressive models, and energy-based models) and discuss the connections between diffusion models and these generative models. Then we review the applications of diffusion models, including computer vision, natural language processing, waveform signal processing, multi-modal modeling, molecular graph generation, time series modeling, and adversarial purification. Furthermore, we propose new perspectives pertaining to the development of generative models. Github: https://github.com/YangLing0818/Diffusion-Models-Papers-Survey-Taxonomy.