@inproceedings{graikos2022diffusion,
author = {Graikos, Alexandros and Malkin, Nikolay and Jojic, Nebojsa and Samaras, Dimitris},
title = {Diffusion Models as Plug-and-Play Priors},
booktitle = {NeurIPS 2022},
year = {2022},
month = {November},
abstract = {We consider the problem of inferring high-dimensional data  in a model that consists of a prior  and an auxiliary differentiable constraint  on  given some additional information . In this paper, the prior is an independently trained denoising diffusion generative model. The auxiliary constraint is expected to have a differentiable form, but can come from diverse sources. The possibility of such inference turns diffusion models into plug-and-play modules, thereby allowing a range of potential applications in adapting models to new domains and tasks, such as conditional generation or image segmentation. The structure of diffusion models allows us to perform approximate inference by iterating differentiation through the fixed denoising network enriched with different amounts of noise at each step. Considering many noised versions of  in evaluation of its fitness is a novel search mechanism that may lead to new algorithms for solving combinatorial optimization problems.},
url = {http://approjects.co.za/?big=en-us/research/publication/diffusion-models-as-plug-and-play-priors/},
}