2020/05/31 by Zhifeng Kong, Kamalika Chaudhuri, Kong, Zhifeng +1 · 1 citation
Computer Science · #Algorithms and Data Compression #Artificial Intelligence in Games #Cellular Automata and Applications #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)
paper · pdf · doi:10.48550/arxiv.2006.00392
openalex publication_date 2020/05/31 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
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.