A Randomized Block Coordinate Iterative Regularized Subgradient Method for High-dimensional Ill-posed Convex Optimization

2019 ◽  
Author(s):  
Harshal Kaushik ◽  
Farzad Yousefian
Keyword(s):  
Convex Optimization ◽  
Subgradient Method ◽  
High Dimensional ◽  
Ill Posed

scholarly journals Distributed Event-Triggered Subgradient Method for Convex Optimization with a Common Constraint Set

2017 ◽  
Vol 50 (1) ◽  
pp. 15319-15324 ◽  
Author(s):  
Yuichi Kajiyama ◽  
Naoki Hayashi ◽  
Shigemasa Takai
Keyword(s):  
Convex Optimization ◽  
Subgradient Method ◽  
Constraint Set ◽  
Event Triggered

scholarly journals Cascaded Compressed Sensing Networks: A Reversible Architecture for Layerwise Learning

2021 ◽  
Author(s):  
weizhi lu ◽  
Mingrui Chen ◽  
kai guo ◽  
Weiyu Li
Keyword(s):  
Compressed Sensing ◽  
High Dimensional ◽  
Layer By Layer ◽  
Network Learning ◽  
Global Target ◽  
Main Challenge ◽  
Ill Posed ◽  
Low Dimensional

Recently, the method that learns networks layer by layer has attracted increasing interest for its ease of analysis. For the method, the main challenge lies in deriving an optimization target for each layer by inversely propagating the global target of the network. The target propagation is an ill-posed problem, due to involving the inversion of nonlinear activations from low-dimensional to high-dimensional spaces. To address the problem, the existing solution is to learn an auxiliary network to specially propagate the target. However, the network lacks stability, and moreover, it leads to higher complexity for network learning. In the letter, we show that target propagation could be achieved by modeling the network's each layer with compressed sensing, without the need of auxiliary networks. Experiments show that the proposed method could achieve better performance than the auxiliary network-based method.<br>

Sign in / Sign up

Export Citation Format

Share Document