Arbitrary Distributions Mapping via SyMOT-Flow: A Flow-based Approach Integrating Maximum Mean Discrepancy and Optimal Transport
This work addresses a crucial challenge in modeling complex data distributions for tasks like sample generation and domain adaptation, representing an incremental improvement by integrating existing techniques in a novel way.
The paper tackles the problem of transforming between two unknown probability distributions using finite samples, introducing SyMOT-Flow, a flow-based model that minimizes symmetric maximum mean discrepancy with optimal transport regularization, resulting in more stable and accurate sample generation validated on low-dimensional examples and high-dimensional medical images.
Finding a transformation between two unknown probability distributions from finite samples is crucial for modeling complex data distributions and performing tasks such as sample generation, domain adaptation and statistical inference. One powerful framework for such transformations is normalizing flow, which transforms an unknown distribution into a standard normal distribution using an invertible network. In this paper, we introduce a novel model called SyMOT-Flow that trains an invertible transformation by minimizing the symmetric maximum mean discrepancy between samples from two unknown distributions, and an optimal transport cost is incorporated as regularization to obtain a short-distance and interpretable transformation. The resulted transformation leads to more stable and accurate sample generation. Several theoretical results are established for the proposed model and its effectiveness is validated with low-dimensional illustrative examples as well as high-dimensional bi-modality medical image generation through the forward and reverse flows.