The Expressive Power of a Class of Normalizing Flow Models
The Expressive Power of a Class of Normalizing Flow Models
复制标题
DOI：
--
复制
发表时间：
2020-05
期刊：
影响因子：
--
通讯作者：
Zhifeng Kong;Kamalika Chaudhuri
中科院分区：
文献类型：
--
作者：
Zhifeng Kong;Kamalika Chaudhuri
标准化流量引起了最近的关注，因为它们允许灵活的生成建模以及易于的可能性计算。尽管已经提出了各种各样的流程模型，但对这些模型的表示能力几乎没有正式的理解。在这项工作中，我们研究了一些基本的归一化流，并严格地建立了其表达能力的界限。我们的结果表明，尽管这些流在一个维度上具有高度表达性，但在较高的维度上，它们的表示功率可能受到限制，尤其是当流量中等深度时。
Normalizing flows have received a great deal of recent attention as they allow flexible generative modeling as well as easy likelihood computation. While a wide variety of flow models have been proposed, there is little formal understanding of the representation power of these models. In this work, we study some basic normalizing flows and rigorously establish bounds on their expressive power. Our results indicate that while these flows are highly expressive in one dimension, in higher dimensions their representation power may be limited, especially when the flows have moderate depth.