Proximal Alternating Direction Network: A Globally Converged Deep Unrolling Framework

11/21/2017
by   Risheng Liu, et al.
0

Deep learning models have gained great success in many real-world applications. However, most existing networks are typically designed in heuristic manners, thus lack of rigorous mathematical principles and derivations. Several recent studies build deep structures by unrolling a particular optimization model that involves task information. Unfortunately, due to the dynamic nature of network parameters, their resultant deep propagation networks do not possess the nice convergence property as the original optimization scheme does. This paper provides a novel proximal unrolling framework to establish deep models by integrating experimentally verified network architectures and rich cues of the tasks. More importantly, we prove in theory that 1) the propagation generated by our unrolled deep model globally converges to a critical-point of a given variational energy, and 2) the proposed framework is still able to learn priors from training data to generate a convergent propagation even when task information is only partially available. Indeed, these theoretical results are the best we can ask for, unless stronger assumptions are enforced. Extensive experiments on various real-world applications verify the theoretical convergence and demonstrate the effectiveness of designed deep models.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/18/2019

Investigating Task-driven Latent Feasibility for Nonconvex Image Modeling

Properly modeling the latent image distributions always plays a key role...
research
09/24/2019

On the Convergence of ADMM with Task Adaption and Beyond

Along with the development of learning and vision, Alternating Direction...
research
12/10/2020

Learning Optimization-inspired Image Propagation with Control Mechanisms and Architecture Augmentations for Low-level Vision

In recent years, building deep learning models from optimization perspec...
research
07/06/2019

Bilevel Integrative Optimization for Ill-posed Inverse Problems

Classical optimization techniques often formulate the feasibility of the...
research
11/18/2017

Learning Aggregated Transmission Propagation Networks for Haze Removal and Beyond

Single image dehazing is an important low-level vision task with many ap...
research
10/09/2018

Learning Converged Propagations with Deep Prior Ensemble for Image Enhancement

Enhancing visual qualities of images plays very important roles in vario...
research
07/17/2019

Underexposed Image Correction via Hybrid Priors Navigated Deep Propagation

Enhancing visual qualities for underexposed images is an extensively con...

Please sign up or login with your details

Forgot password? Click here to reset