A Unifying Tutorial on Approximate Message Passing
A Unifying Tutorial on Approximate Message Passing
复制标题
DOI：
10.1561/2200000092
复制
发表时间：
2022-01-01
影响因子：
32.8
通讯作者：
Samworth, Richard J.
中科院分区：
文献类型：
--
作者：
Feng, Oliver Y.;Venkataramanan, Ramji;Samworth, Richard J.
在过去的十年左右，近似消息传递（AMP）算法已经成为非常流行的各种结构化的高维统计问题。虽然这些技术的起源可以追溯到统计物理学文献中的信念传播的概念，我们在这项工作中的目标是从统计的角度提出AMP的主要思想，并说明AMP框架的功能和灵活性。沿着的方式，我们加强和统一的许多结果，在现有的文献。
Over the last decade or so, Approximate Message Passing (AMP) algorithms have become extremely popular in various structured high-dimensional statistical problems. Although the origins of these techniques can be traced back to notions of belief propagation in the statistical physics literature, our goals in this work are to present the main ideas of AMP from a statistical perspective and to illustrate the power and flexibility of the AMP framework. Along the way, we strengthen and unify many of the results in the existing literature.