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 0907.0914cs.ITcond-mat.dis-nnmath.ITmath.STstat.TH
收藏

基于Lp范数最小化的压缩感知重构极限

A typical reconstruction limit of compressed sensing based on Lp-norm minimization

Y. Kabashima, T. Wadayama, T. Tanaka

摘要:

我们考虑的问题,重建$N$维连续向量$\bx$从$P$约束下产生的线性变换的假设下,$\bx$的非零元素的数量通常是有限的$ρN$($0\le ρ\le 1$)。这种类型的问题可以通过最小化成本函数的$L_p$ -范数$||\bx||_p=\lim_{ε\to +0}\sum_{i=1}^N |x_i|^{p+ε}$,在适当的条件下的约束。对于几个$p$,我们评估了典型情况极限$α_c(ρ)$,它代表$α=P/N$和$ρ$之间的关键关系,用于通过在极限$N,P \to \infty$中的典型情况下最小化并利用复制方法保持$α$有限来成功重建原始向量。对于$p=1$,$α_c(ρ)$比其最差情况对应物小得多,这是由现有的信息论文献严格推导出来的。

英文摘要:

We consider the problem of reconstructing an $N$-dimensional continuous vector $\bx$ from $P$ constraints which are generated by its linear transformation under the assumption that the number of non-zero elements of $\bx$ is typically limited to $ρN$ ($0\le ρ\le 1$). Problems of this type can be solved by minimizing a cost function with respect to the $L_p$-norm $||\bx||_p=\lim_{ε\to +0}\sum_{i=1}^N |x_i|^{p+ε}$, subject to the constraints under an appropriate condition. For several $p$, we assess a typical case limit $α_c(ρ)$, which represents a critical relation between $α=P/N$ and $ρ$ for successfully reconstructing the original vector by minimization for typical situations in the limit $N,P \to \infty$ with keeping $α$ finite, utilizing the replica method. For $p=1$, $α_c(ρ)$ is considerably smaller than its worst case counterpart, which has been rigorously derived by existing literature of information theory.

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。

↑