Leveraging Variational Autoencoders for Parameterized MMSE Estimation

arXiv:2307.05352v2 Announce Type: replace-cross
Abstract: In this manuscript, we propose to use a variational autoencoder-based framework for parameterizing a conditional linear minimum mean squared error estimator. The variational autoencoder models the underlying unknown data distribution as conditionally Gaussian, yielding the conditional first and second moments of the estimand, given a noisy observation. The derived estimator is shown to approximate the minimum mean squared error estimator by utilizing the variational autoencoder as a generative prior for the estimation problem. We propose …

abstract arxiv autoencoder autoencoders cs.it data distribution eess.sp error estimator framework linear math.it mean moments observation stat.ml type variational autoencoders