Recovering a Hidden Community Beyond the Spectral Limit in O(|E|log*|V|) Time

Recovering a Hidden Community Beyond the Spectral Limit in O(|E|log*|V|) Time
复制标题

在 O(|E|log*|V|) 时间内恢复超出光谱限制的隐藏社区

DOI:
--
复制
发表时间:
2015
期刊:
arXiv.org
影响因子:
--
通讯作者:
Jiaming Xu
Jiaming Xu
中科院分区:
--
文献类型:
--
作者:
B. Hajek;Yihong Wu;Jiaming Xu

文献摘要

参考文献

被引文献

相似文献

考虑具有参数 $n、K、p、$ 和 $q$ 的一个社区的随机块模型: $n$ 个顶点中的 $K$ 在社区中;如果两个顶点都在社区中,则它们通过边以 $p$ 的概率连接,否则以 $q$ 的概率连接,其中 $p > q > 0$ 且 $p/q$ 被假定为有界。如果错误分类顶点的平均数量为 $o(K)$ 作为 $n o infty$,则基于对图 $G=(V,E)$ 的观察的估计器据说会实现弱恢复。有效信噪比 $lambda=K^2(p-q)^2/((n-K)q) 发挥着关键作用。$在 $K=Theta(n)$ 体系中,如果 $lambda o infty$,简单的度阈值算法可以在 $O(|E|)$ 时间内实现弱恢复,这与弱恢复的信息论可能性相一致。 本文的主要重点是亚线性状态 $K=o(n)$ 和 $np = n^{o(1)}$ 中的弱恢复。$ 结果表明,弱恢复是由运行 $log^ast(n)+O(1) $ 次迭代的置信传播算法提供的,如果 $lambda > 1/e,$ 总时间复杂度为 $O(|E| log^*n)$。相反,如果 $lambda leq 1/e$,交互作用半径 $t$ 满足 $t = o(frac{log n}{log(2+np)})$ 的局部算法无法渐近胜过普通随机猜测。通过分析对应于将幂迭代应用于图的非回溯矩阵的线性消息传递算法,我们提供证据表明,如果 $lambda leq 1.$,谱方法无法提供弱恢复。
The stochastic block model for one community with parameters $n, K, p,$ and $q$ is considered: $K$ out of $n$ vertices are in the community; two vertices are connected by an edge with probability $p$ if they are both in the community and with probability $q$ otherwise, where $p > q > 0$ and $p/q$ is assumed to be bounded. An estimator based on observation of the graph $G=(V,E)$ is said to achieve weak recovery if the mean number of misclassified vertices is $o(K)$ as $n o infty$. A critical role is played by the effective signal-to-noise ratio $lambda=K^2(p-q)^2/((n-K)q).$ In the regime $K=Theta(n)$, a na"{i}ve degree-thresholding algorithm achieves weak recovery in $O(|E|)$ time if $lambda o infty$, which coincides with the information theoretic possibility of weak recovery. The main focus of the paper is on weak recovery in the sublinear regime $K=o(n)$ and $np = n^{o(1)}.$ It is shown that weak recovery is provided by a belief propagation algorithm running for $log^ast(n)+O(1) $ iterations, if $lambda > 1/e,$ with the total time complexity $O(|E| log^*n)$. Conversely, no local algorithm with radius $t$ of interaction satisfying $t = o(frac{log n}{log(2+np)})$ can asymptotically outperform trivial random guessing if $lambda leq 1/e.$ By analyzing a linear message-passing algorithm that corresponds to applying power iteration to the non-backtracking matrix of the graph, we provide evidence to suggest that spectral methods fail to provide weak recovery if $lambda leq 1.$
DOI: 10.1214/15-aap1145
发表时间: 2013-09
期刊: --
影响因子: --
作者:
Elchanan Mossel;Joe Neeman;A. Sly
通讯作者: Elchanan Mossel;Joe Neeman;A. Sly
通过消息传递进行子矩阵定位
DOI: --
发表时间: 2018
影响因子: 6
作者:
Hajek, Bruce;Wu, Yihong;Xu, Jiaming
通讯作者: Xu, Jiaming