On the Optimization Landscape of Maximum Mean Discrepancy

arXiv:2110.13452v2 Announce Type: replace
Abstract: Generative models have been successfully used for generating realistic signals. Because the likelihood function is typically intractable in most of these models, the common practice is to use "implicit" models that avoid likelihood calculation. However, it is hard to obtain theoretical guarantees for such models. In particular, it is not understood when they can globally optimize their non-convex objectives. Here we provide such an analysis for the case of Maximum Mean Discrepancy (MMD) learning of …

abstract arxiv cs.lg function generative generative models however landscape likelihood maximum mean optimization practice stat.ml type