资源论文Approximate message passing for amplitude based optimization

Approximate message passing for amplitude based optimization

2020-03-11 | |  66 |   61 |   0

Abstract

We consider an 图片.png-regularized non-convex optimization problem for recovering signals from their noisy phaseless observations. We design and study the performance of a message passing algorithm that aims to solve this optimization problem. We consider the asymptotic setting m, n 图片.png∞ , m/n 图片.png  δ and obtain sharp performance bounds, where m is the number of measurements and n is the signal dimension. We show that for complex signals the algorithm can perform accurate recovery with only m = 图片.png measurements. Also, we provide sharp analysis on the sensitivity of the algorithm to noise. We highlight the following facts about our message passing algorithm: (i) Adding 图片.png regularization to the non-convex loss function can be beneficial even in the noiseless setting; (ii) spectral initialization has marginal impact on the performance of the algorithm.

上一篇:Extreme Learning to Rank via Low Rank Assumption

下一篇:Pathwise Derivatives Beyond the Reparameterization Trick

用户评价
全部评价

热门资源

  • A Mathematical Mo...

    Direct democracy, where each voter casts one vo...

  • Stratified Strate...

    In this paper we introduce Stratified Strategy ...

  • Learning to Predi...

    Much of model-based reinforcement learning invo...

  • dynamical system ...

    allows to preform manipulations of heavy or bul...

  • The Variational S...

    Unlike traditional images which do not offer in...