Regarding two conjectures on clique and biclique partitions
arXiv:2005.02529
Abstract
For a graph , let denote the minimum number of cliques of needed to cover the edges of exactly once. Similarly, let denote the minimum number of bicliques (i.e. complete bipartite subgraphs of ) needed to cover each edge of exactly times. We consider two conjectures -- one regarding the maximum possible value of (due to de Caen, Erdős, Pullman and Wormald) and the other regarding (due to de Caen, Gregory and Pritikin). We disprove the first, obtaining improved lower and upper bounds on , and we prove an asymptotic version of the second, showing that .