Diffusion probabilistic models (DPMs) are emerging powerful generative models. Despite their high-quality generation performance, DPMs still suffer from their slow sampling as they generally need hundreds or thousands of sequential function evaluations (steps) of large neural networks to draw a sample. Sampling from DPMs can be viewed alternatively as solving the corresponding diffusion ordinary differential equations (ODEs). In this work, we propose an exact formulation of the solution of diffusion ODEs. The formulation analytically computes the linear part of the solution, rather than leaving all terms to black-box ODE solvers as adopted in previous works. By applying change-of-variable, the solution can be equivalently simplified to an exponentially weighted integral of the neural network. Based on our formulation, we propose DPM-Solver, a fast dedicated high-order solver for diffusion ODEs with the convergence order guarantee. DPM-Solver is suitable for both discrete-time and continuous-time DPMs without any further training. Experimental results show that DPM-Solver can generate high-quality samples in only 10 to 20 function evaluations on various datasets. We achieve 4.70 FID in 10 function evaluations and 2.87 FID in 20 function evaluations on the CIFAR10 dataset, and a $4\sim 16\times$ speedup compared with previous state-of-the-art training-free samplers on various datasets.
翻译:聚变概率模型(DPMs)正在形成强大的基因化模型。尽管其生成性能是高质量的,但DPMs仍然受到缓慢的抽样,因为它们通常需要数百或数千个大型神经网络的连续功能评价(步骤)来提取样本。从DPMs取样可以被看成是解决相应的扩散普通差异方程式(ODEs)的解决方案。在这项工作中,我们建议精确地制定扩散源码的解决方案。这种配方在分析上计算了解决方案的线性部分,而不是将所有条件留给先前工作中采用的黑盒 ODE 解算器。通过应用变换式,解决方案可以等同于神经网络的指数加权组合(步骤)。基于我们的配方,我们建议DPM-Solver(DPM-Solver)是一个快速专用的高级解析器,用于使用聚合序列保证的传播源码。DPM-Sver(DPM-Sver)适合于不连续的DPMM(线性解算)部分,而无需任何进一步培训。实验结果表明,DPM-Solderver(DPM-70)能够生成高品质样本样本样本,而仅10至20次的样本样本样本样本样本样本样本样本,而进行数据评估。