Abstract: Plug-and-play priors (PnP) is a powerful framework for regularizing imaging problems by using advanced denoisers, such as those based on deep neural networks, within an iterative algorithm. Recent experimental evidence suggests that PnP algorithms achieve state-of-the-art performance in a range of imaging applications. In this talk, we introduce a new online PnP algorithm that use only a subset of measurements at every iteration. This makes the algorithm scalable to very large datasets. We present a new theoretical convergence analysis, for both batch and online variants of PnP, for denoisers that do not necessarily correspond to proximal operators. We also present simulations illustrating the applicability of the algorithm to image reconstruction in diffraction tomography. Our results in this paper have the potential to expand the applicability of the PnP framework to very large and redundant datasets. This talk is based on the recent manuscript: https://arxiv.org/abs/1809.04693.