论文标题
子空间扩散生成模型
Subspace Diffusion Generative Models
论文作者
论文摘要
基于得分的模型通过通过高维扩散过程将噪声映射到数据（反之亦然）来生成样品。我们质疑是否有必要在高维度上运行整个过程，并带来所有不便。取而代之的是，随着数据分布朝向噪声，我们限制了通过投影扩散到子空间。当应用于最新模型时，我们的框架同时提高了样本质量 - 在无条件的CIFAR-10上达到了2.17的FID，并降低了相同数量的DeNoising步骤的计算成本。我们的框架与连续的时间扩散完全兼容，并保留其柔性功能，包括精确的对数可能和可控的生成。代码可从https://github.com/bjing2016/subspace-diffusion获得。
Score-based models generate samples by mapping noise to data (and vice versa) via a high-dimensional diffusion process. We question whether it is necessary to run this entire process at high dimensionality and incur all the inconveniences thereof. Instead, we restrict the diffusion via projections onto subspaces as the data distribution evolves toward noise. When applied to state-of-the-art models, our framework simultaneously improves sample quality -- reaching an FID of 2.17 on unconditional CIFAR-10 -- and reduces the computational cost of inference for the same number of denoising steps. Our framework is fully compatible with continuous-time diffusion and retains its flexible capabilities, including exact log-likelihoods and controllable generation. Code is available at https://github.com/bjing2016/subspace-diffusion.
