Async-RED: A Provably Convergent Asynchronous Block Parallel Stochastic Method using Deep Denoising Priors
arXiv:2010.01446
Abstract
Regularization by denoising (RED) is a recently developed framework for solving inverse problems by integrating advanced denoisers as image priors. Recent work has shown its state-of-the-art performance when combined with pre-trained deep denoisers. However, current RED algorithms are inadequate for parallel processing on multicore systems. We address this issue by proposing a new asynchronous RED (ASYNC-RED) algorithm that enables asynchronous parallel processing of data, making it significantly faster than its serial counterparts for large-scale inverse problems. The computational complexity of ASYNC-RED is further reduced by using a random subset of measurements at every iteration. We present complete theoretical analysis of the algorithm by establishing its convergence under explicit assumptions on the data-fidelity and the denoiser. We validate ASYNC-RED on image recovery using pre-trained deep denoisers as priors.
References in corpus (12)
- Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
- HOGWILD!: A Lock-Free Approach to Parallelizing Stochastic Gradient Descent
- Optimal Distributed Online Prediction using Mini-Batches
- Asynchronous Parallel Stochastic Gradient for Nonconvex Optimization
- Plug and play methods for magnetic resonance imaging (long version)
- An Online Plug-and-Play Algorithm for Regularized Image Reconstruction
- Plug-and-Play Methods Provably Converge with Properly Trained Denoisers
- SIMBA: Scalable Inversion in Optical Tomography using Deep Denoising Priors
- Plug-in Estimation in High-Dimensional Linear Inverse Problems: A Rigorous Analysis
- Asynchronous Coordinate Descent under More Realistic Assumptions
- Regularized Fourier Ptychography using an Online Plug-and-Play Algorithm
- DeepRED: Deep Image Prior Powered by RED