arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1305.0047stat.ML
收藏

字典LASSO:线性变换下保证稀疏恢复

Dictionary LASSO: Guaranteed Sparse Recovery under Linear Transformation

Ji Liu, Lei Yuan, Jieping Ye

摘要:

我们考虑以下信号恢复问题:给定测量矩阵$Φ\in \mathbb{R}^{n\times p}$和由$c = Φθ^* + ε$构造的噪声观测向量$c\in \mathbb{R}^{n}$,其中$ε\in \mathbb{R}^{n}$是其条目遵循i.i.d.如果$Dθ^*$是稀疏的,如何恢复信号$θ^*$ {线性变换下的\rca} $D\in\mathbb{R}^{m\times p}$?一个自然的方法,使用凸优化是解决以下问题:$$\min_θ {1\over 2}\|Φθ- c\|^2 + λ\|Dθ\|_1.$$本文提供了一个上限的估计误差,并显示了一致性的性质,这种方法假设设计矩阵$Φ$是一个高斯随机矩阵。具体地说,我们证明了:1)在无噪声情况下,如果$D$的条件数有界,并且测量数$n\geq Ω(s\log(p))$(其中$s$是稀疏数),则真解可以以高概率恢复;以及2)在有噪声的情况下,如果$D$的条件数有界并且测量比$s\log(p)$增加得快,即,$s\log(p)=o(n)$,当$p$和$s$趋于无穷大时,估计误差以概率1收敛到零。我们的结果是一致的特殊情况$D=\bold{I}_{p\times p}$(相当于LASSO)和改善现有的分析。$D$的条件数在我们的分析中起着关键作用。我们考虑了两种情况下的条件数,包括融合LASSO和随机图:融合LASSO情况下的条件数由常数限定,而随机图情况下的条件数以高概率限定,如果$m\over p$(即,$#text{edge}\over #text{vertex}$)大于某一常数。数值模拟结果与我们的理论结果是一致的。

英文摘要:

We consider the following signal recovery problem: given a measurement matrix $Φ\in \mathbb{R}^{n\times p}$ and a noisy observation vector $c\in \mathbb{R}^{n}$ constructed from $c = Φθ^* + ε$ where $ε\in \mathbb{R}^{n}$ is the noise vector whose entries follow i.i.d. centered sub-Gaussian distribution, how to recover the signal $θ^*$ if $Dθ^*$ is sparse {\rca under a linear transformation} $D\in\mathbb{R}^{m\times p}$? One natural method using convex optimization is to solve the following problem: $$\min_θ {1\over 2}\|Φθ- c\|^2 + λ\|Dθ\|_1.$$ This paper provides an upper bound of the estimate error and shows the consistency property of this method by assuming that the design matrix $Φ$ is a Gaussian random matrix. Specifically, we show 1) in the noiseless case, if the condition number of $D$ is bounded and the measurement number $n\geq Ω(s\log(p))$ where $s$ is the sparsity number, then the true solution can be recovered with high probability; and 2) in the noisy case, if the condition number of $D$ is bounded and the measurement increases faster than $s\log(p)$, that is, $s\log(p)=o(n)$, the estimate error converges to zero with probability 1 when $p$ and $s$ go to infinity. Our results are consistent with those for the special case $D=\bold{I}_{p\times p}$ (equivalently LASSO) and improve the existing analysis. The condition number of $D$ plays a critical role in our analysis. We consider the condition numbers in two cases including the fused LASSO and the random graph: the condition number in the fused LASSO case is bounded by a constant, while the condition number in the random graph case is bounded with high probability if $m\over p$ (i.e., $#text{edge}\over #text{vertex}$) is larger than a certain constant. Numerical simulations are consistent with our theoretical results.

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。

↑