Universality of Approximate Message Passing Algorithms
Probability
2020-06-02 v2 Information Theory
Mathematical Physics
math.IT
math.MP
Optimization and Control
Statistics Theory
Statistics Theory
Abstract
We consider a broad class of Approximate Message Passing (AMP) algorithms defined as a Lipschitzian functional iteration in terms of an random symmetric matrix . We establish universality in noise for this AMP in the -limit and validate this behavior in a number of AMPs popularly adapted in compressed sensing, statistical inferences, and optimizations in spin glasses.
Keywords
Cite
@article{arxiv.2003.10431,
title = {Universality of Approximate Message Passing Algorithms},
author = {Wei-Kuo Chen and Wai-Kit Lam},
journal= {arXiv preprint arXiv:2003.10431},
year = {2020}
}
Comments
43 pages, minor revision in the introduction