The Average-Case Complexity of Counting Cliques in Erdos-Renyi Hypergraphs

From MaRDI portal
Publication:6315907

DOI10.1137/20M1316044arXiv1903.08247MaRDI QIDQ6315907FDOQ6315907


Authors: Enric Boix-Adserà, Matthew D. Brennan, Guy Bresler Edit this on Wikidata


Publication date: 19 March 2019

Abstract: We consider the problem of counting k-cliques in s-uniform Erdos-Renyi hypergraphs G(n,c,s) with edge density c, and show that its fine-grained average-case complexity can be based on its worst-case complexity. We prove the following: 1. Dense Erdos-Renyi graphs and hypergraphs: Counting k-cliques on G(n,c,s) with k and c constant matches its worst-case time complexity up to a mathrmpolylog(n) factor. Assuming randomized ETH, it takes nOmega(k) time to count k-cliques in G(n,c,s) if k and c are constant. 2. Sparse Erdos-Renyi graphs and hypergraphs: When c=Theta(nalpha), we give several algorithms exploiting the sparsity of G(n,c,s) that are faster than the best known worst-case algorithms. Complementing this, based on a fine-grained worst-case assumption, our results imply a different average-case phase diagram for each fixed alpha depicting a tradeoff between a runtime lower bound and k. Surprisingly, in the hypergraph case (sge3), these lower bounds are tight against our algorithms exactly when c is above the ErdH{o}s-R'{e}nyi k-clique percolation threshold. This is the first worst-case-to-average-case hardness reduction for a problem on ErdH{o}s-R'{e}nyi hypergraphs that we are aware of. We also give a variant of our result for computing the parity of the k-clique count that tolerates higher error probability.













This page was built for publication: The Average-Case Complexity of Counting Cliques in Erdos-Renyi Hypergraphs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6315907)