Feature Likelihood Divergence: Evaluating the Generalization of Generative Models Using Samples

arXiv:2302.04440v4 Announce Type: replace
Abstract: The past few years have seen impressive progress in the development of deep generative models capable of producing high-dimensional, complex, and photo-realistic data. However, current methods for evaluating such models remain incomplete: standard likelihood-based metrics do not always apply and rarely correlate with perceptual fidelity, while sample-based metrics, such as FID, are insensitive to overfitting, i.e., inability to generalize beyond the training set. To address these limitations, we propose a new metric called the Feature …

arxiv cs.cv cs.lg divergence feature generative generative models likelihood samples type