核心思路：噪声与去噪

这部分内容有帮助吗？

参考文献

Deep Unsupervised Learning using Nonequilibrium Thermodynamics, Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, and Stefano Ermon, 2015 Proceedings of the 32nd International Conference on Machine Learning (ICML), Vol. 37 (PMLR (Proceedings of Machine Learning Research)) DOI: 10.55989/ah-3 - 这篇是首次提出用于生成任务的扩散概率模型概念的论文，为后续进展奠定了理论基础。
Denoising Diffusion Probabilistic Models, Jonathan Ho, Ajay Jain, and Pieter Abbeel, 2020 Advances in Neural Information Processing Systems 33 (NeurIPS) DOI: 10.48550/arXiv.2006.11239 - 本文改进并推广了去噪扩散概率模型 (DDPM) 框架，提供了更简单的训练目标，并在图像生成方面展示了优异效果，使扩散模型得到广泛应用。
Score-Based Generative Modeling through Stochastic Differential Equations, Yang Song, Jascha Sohl-Dickstein, Diederik P. Kingma, Abhishek Kumar, Stefano Ermon, Ben Poole, 2020 International Conference on Learning Representations (ICLR) DOI: 10.48550/arXiv.2011.13456 - 这项研究将去噪扩散概率模型与基于分数的生成模型统一在一个连续时间随机微分方程框架下，显著推动了这些模型的理论理解和实际应用。