Answers are grounded in this paper and cite their sources.
In recent years, there have been significant advances in the use of deep learning methods in inverse problems such as denoising, compressive sensing, inpainting, and super-resolution. While this line of works has predominantly been driven by practical algorithms and experiments, it has also given rise to a variety of intriguing theoretical problems. In this paper, we survey some of the prominent theoretical developments in this line of works, focusing in particular on generative priors, untrained neural network priors, and unfolding algorithms. In addition to summarizing existing results in these topics, we highlight several ongoing challenges and open problems.
No takes yet. Share an insight, caveat, or question.
Scarlett et al. (2022) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: