论文标题
组成逆问题的归一流流
Composing Normalizing Flows for Inverse Problems
论文作者
论文摘要
考虑到归一流流量的逆问题，我们希望估算观测值条件的基础信号的分布。我们将这个问题作为对预训练的无条件流量模型有条件推断的任务。我们首先确定这对于大量流量模型来说很难计算。在此激励的情况下，我们为近似推断提出了一个框架，该框架将目标条件估计为两个流模型的组成。这种表述会导致稳定的变异推理训练程序，以避免对抗性训练。我们的方法在各种反问题上进行了评估，并显示出可产生具有不确定性定量的高质量样品。我们进一步证明，我们的方法可以摊销以零击中的推断。
Given an inverse problem with a normalizing flow prior, we wish to estimate the distribution of the underlying signal conditioned on the observations. We approach this problem as a task of conditional inference on the pre-trained unconditional flow model. We first establish that this is computationally hard for a large class of flow models. Motivated by this, we propose a framework for approximate inference that estimates the target conditional as a composition of two flow models. This formulation leads to a stable variational inference training procedure that avoids adversarial training. Our method is evaluated on a variety of inverse problems and is shown to produce high-quality samples with uncertainty quantification. We further demonstrate that our approach can be amortized for zero-shot inference.
