Kernelised Normalising Flows

arXiv:2307.14839v4 Announce Type: replace-cross
Abstract: Normalising Flows are non-parametric statistical models characterised by their dual capabilities of density estimation and generation. This duality requires an inherently invertible architecture. However, the requirement of invertibility imposes constraints on their expressiveness, necessitating a large number of parameters and innovative architectural designs to achieve good results. Whilst flow-based models predominantly rely on neural-network-based transformations for expressive designs, alternative transformation methods have received limited attention. In this work, we present Ferumal flow, a novel kernelised …

abstract architecture arxiv capabilities constraints cs.lg designs flow good however non-parametric parameters parametric replace results statistical stat.ml type