We provide attainable analytical tools to estimate the error of flow-based generative models under the Wasserstein metric and to establish the optimal sampling iteration complexity bound with respect to dimension as $O(\sqrt{d})$. We show this error can be explicitly controlled by two parts: the Lipschitzness of the push-forward maps of the backward flow which scales independently of the dimension; and a local discretization error scales $O(\sqrt{d})$ in terms of dimension. The former one is related to the existence of Lipschitz changes of variables induced by the (heat) flow. The latter one consists of the regularity of the score function in both spatial and temporal directions. These assumptions are valid in the flow-based generative model associated with the Föllmer process and $1$-rectified flow under the Gaussian tail assumption. As a consequence, we show that the sampling iteration complexity grows linearly with the square root of the trace of the covariance operator, which is related to the invariant distribution of the forward process.
翻译:我们提供了可实现的解析工具来估计基于流的生成模型在Wasserstein度量下的误差,并建立关于维度$d$的最优采样迭代复杂度界$O(\sqrt{d})$。我们证明该误差可被两部分显式控制:反向流推前映射的Lipschitz性(其缩放与维度无关);以及局部离散化误差在维度上以$O(\sqrt{d})$缩放。前者与(热)流诱导的Lipschitz变量变换的存在性相关。后者包含分数函数在空间和时间方向上的正则性。这些假设在满足高斯尾分布的Föllmer过程与1-整流流相关的流基生成模型中成立。因此,我们证明采样迭代复杂度随协方差算子迹的平方根线性增长,这与前向过程的不变分布相关。