Stochastic Localization via Iterative Posterior Sampling

arXiv:2402.10758v1 Announce Type: cross
Abstract: Building upon score-based learning, new interest in stochastic localization techniques has recently emerged. In these models, one seeks to noise a sample from the data distribution through a stochastic process, called observation process, and progressively learns a denoiser associated to this dynamics. Apart from specific applications, the use of stochastic localization for the problem of sampling from an unnormalized target density has not been explored extensively. This work contributes to fill this gap. We consider …

abstract applications arxiv building cs.lg data distribution dynamics iterative localization noise observation posterior process sample sampling stat.co stat.ml stochastic stochastic process through type via