Improved sampling via learned diffusions

arXiv:2307.01198v2 Announce Type: replace
Abstract: Recently, a series of papers proposed deep learning-based approaches to sample from target distributions using controlled diffusion processes, being trained only on the unnormalized target densities without access to samples. Building on previous work, we identify these approaches as special cases of a generalized Schr\"odinger bridge problem, seeking a stochastic evolution between a given prior distribution and the specified target. We further generalize this framework by introducing a variational formulation based on divergences between path …

abstract access arxiv bridge building cases cs.lg deep learning diffusion generalized identify math.oc math.pr papers processes replace sample samples sampling series stat.ml stochastic type via work