arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2111.03237cs.ITcs.LGmath.IT
收藏

面向广义线性逆问题的最优传感矩阵设计

Towards Designing Optimal Sensing Matrices for Generalized Linear Inverse Problems

Columbia University(哥伦比亚大学);Institute of Computational Mathematics and Scientific/Engineering Computing, Academy of Mathematics and Systems Science, Chinese Academy of Sciences(中国科学院数学与系统科学研究院计算数学与科学工程计算研究所)

Junjie Ma, Ji Xu, Arian Maleki

AI总结:

本文研究广义线性逆问题中传感矩阵对 EP 恢复性能的影响,提出谱尖峰性度量,揭示其作用取决于非线性函数,并据此给出相位恢复与1比特压缩感知中的最优矩阵设计启示。

摘要:

我们考虑逆问题 $\mathbf{y}=f(\mathbf{Ax})$,其中 $\mathbf{x}\in\mathbb{R}^n$ 是感兴趣的信号,$\mathbf{A}$ 是传感矩阵,$f$ 是非线性函数,$\mathbf{y}\in\mathbb{R}^m$ 是测量向量。在许多应用中,我们在设计传感矩阵 $\mathbf{A}$ 时具有一定程度的自由度;在这种情况下,可以优化 $\mathbf{A}$ 以获得更好的重建性能。作为迈向最优设计的第一步,理解传感矩阵对从 $\mathbf{y}$ 恢复 $\mathbf{x}$ 的难度有何影响至关重要。本文研究最成功的恢复方法之一——期望传播(EP)算法——的性能。我们为 $\mathbf{A}$ 的谱定义了“尖峰性”概念,并证明该度量对 EP 性能的重要性。我们表明,更尖峰的谱究竟是损害还是有助于恢复性能取决于 $f$。基于所提框架,我们能够证明:在相位恢复问题中,谱更尖峰的矩阵更适合 EP;而在 1 比特压缩感知问题中,尖峰性较低的谱会带来更好的性能。我们的结果统一并实质性地推广了比较高斯矩阵与正交矩阵的现有结果,并为设计最优传感系统提供了平台。

英文摘要:

We consider an inverse problem $\mathbf{y}= f(\mathbf{Ax})$, where $\mathbf{x}\in\mathbb{R}^n$ is the signal of interest, $\mathbf{A}$ is the sensing matrix, $f$ is a nonlinear function and $\mathbf{y} \in \mathbb{R}^m$ is the measurement vector. In many applications, we have some level of freedom to design the sensing matrix $\mathbf{A}$, and in such circumstances we could optimize $\mathbf{A}$ to achieve better reconstruction performance. As a first step towards optimal design, it is important to understand the impact of the sensing matrix on the difficulty of recovering $\mathbf{x}$ from $\mathbf{y}$. In this paper, we study the performance of one of the most successful recovery methods, i.e., the expectation propagation (EP) algorithm. We define a notion of spikiness for the spectrum of $\bmmathbfA}$ and show the importance of this measure for the performance of EP. We show that whether a spikier spectrum can hurt or help the recovery performance depends on $f$. Based on our framework, we are able to show that, in phase-retrieval problems, matrices with spikier spectrums are better for EP, while in 1-bit compressed sensing problems, less spiky spectrums lead to better performance. Our results unify and substantially generalize existing results that compare Gaussian and orthogonal matrices, and provide a platform towards designing optimal sensing systems.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑