Selecting the Best in GANs Family: a Post Selection Inference Framework
Abstract
"Which Generative Adversarial Networks (GANs) generates the most plausible images?" has been a frequently asked question among researchers. To address this problem, we first propose an incomplete U-statistics estimate of maximum mean discrepancy MMDinc to measure the distribution discrepancy between generated and real images. MMDinc enjoys the advantages of asymptotic normality, computation efficiency, and model agnosticity. We then propose a GANs analysis framework to select and test the "best" member in GANs family using the Post Selection Inference (PSI) with MMDinc. In the experiments, we adopt the proposed framework on 7 GANs variants and compare their MMDinc scores.
Turn this paper into a lesson
ArcXiv compiles a structured reading guide from this paper's metadata: plain-English importance, contributions, prerequisite concepts, which sections to read first, flashcards, and a quiz. Grounded in the abstract, never invented.