Sum-of-Squares Lower Bounds for Densest k-Subgraph
Sum-of-Squares Lower Bounds for Densest k-Subgraph
复制标题
最稠 k 子图的平方和下界
DOI:
10.1145/3564246.3585221
复制
发表时间:
2023
期刊:
影响因子:
--
通讯作者:
Xu, Jeff
中科院分区:
文献类型:
--
作者:
Jones, Chris;Potechin, Aaron;Rajendran, Goutham;Xu, Jeff
Given a graph and an integerk, Densestk-Subgraph is the algorithmic task of finding the subgraph onkvertices with the maximum number of edges. This is a fundamental problem that has been subject to intense study for decades, with applications spanning a wide variety of fields. The state-of-the-art algorithm is anO(n1/4 +)-factor approximation (for any > 0) due to Bhaskara et al. [STOC ’10]. Moreover, the so-calledlog-density frameworkpredicts that this is optimal, i.e. it is impossible for an efficient algorithm to achieve anO(n1/4 −)-factor approximation. In the average case, Densestk-Subgraph is a prototypical noisy inference task which is conjectured to exhibit astatistical-computational gap.In this work, we provide the strongest evidence yet of hardness for Densestk-Subgraph by showing matching lower bounds against the powerful Sum-of-Squares (SoS) algorithm, a meta-algorithm based on convex programming that achieves state-of-art algorithmic guarantees for many optimization and inference problems. Fork≤n1/2, we obtain a degreenδSoS lower bound for the hard regime as predicted by the log-density framework.To show this, we utilize the modern framework for proving SoS lower bounds on average-case problems pioneered by Barak et al. [FOCS ’16]. A key issue is that small denser-than-average subgraphs in the input will greatly affect the value of the candidate pseudoexpectation operator around the subgraph. To handle this challenge, we devise a novel matrix factorization scheme based on thepositive minimum vertex separator. We then prove an intersection tradeoff lemma to show that the error terms when using this separator are indeed small.
登录
查看更多内容
DOI:
10.1145/3055399.3055438
发表时间:
2016-10
期刊:
Proceedings of the 49th Annual ACM SIGACT Symposium on Theory of Computing
影响因子:
--
作者:
Pravesh Kothari;Raghu Meka;P. Raghavendra
通讯作者:
Pravesh Kothari;Raghu Meka;P. Raghavendra
DOI:
--
发表时间:
2017
期刊:
arXiv.org
影响因子:
--
作者:
Pravesh Kothari;David Steurer
通讯作者:
David Steurer
DOI:
--
发表时间:
2020-05
期刊:
--
影响因子:
--
作者:
Matthew Brennan;Guy Bresler
通讯作者:
Matthew Brennan;Guy Bresler
DOI:
10.1145/3055399.3055412
发表时间:
2016-11
期刊:
Proceedings of the 49th Annual ACM SIGACT Symposium on Theory of Computing
影响因子:
--
作者:
Pasin Manurangsi
通讯作者:
Pasin Manurangsi
影响因子:
1.9
作者:
Ames, Brendan P. W.
通讯作者:
Ames, Brendan P. W.