We present high quality image synthesis results using diffusion probabilistic models, a class of latent variable models inspired by considerations from nonequilibrium thermodynamics. Our best results are obtained by training on a weighted variational bound designed according to a novel connection between diffusion probabilistic models and denoising score matching with Langevin dynamics, and our models naturally admit a progressive lossy decompression scheme that can be interpreted as a generalization of autoregressive decoding. On the unconditional CIFAR10 dataset, we obtain an Inception score of 9.46 and a state-of-the-art FID score of 3.17. On 256x256 LSUN, we obtain sample quality similar to ProgressiveGAN. Our implementation is available at https://github.com/hojonathanho/diffusion

本文使用受非平衡热力学考虑的潜变量模型——扩散概率模型，提出了高质量的图像合成结果。通过根据扩散概率模型和Langevin动力学的去噪得分匹配之间的新颖联系设计加权变分界限进行训练，获得了最佳结果；此外，我们的模型自然地采用渐进式有损解压缩方案，可以解释为自回归解码的一般化。在无条件的CIFAR10数据集上，我们获得了9.46的Inception得分和3.17的最先进的FID得分。在256x256 LSUN上，我们获得了与ProgressiveGAN相似的样本质量。

去噪扩散概率模型