Constrained multilinear detection for faster functional motif discovery

Abstract

The Graph Motif problem asks whether a given multiset of colors appears on a connected subgraph of a vertex-colored graph. The fastest known parameterized algorithm for this problem is based on a reduction to the k-Multilinear Detection (k-MlD) problem: the detection of multilinear terms of total degree k in polynomials presented as circuits. We revisit k-MlD and define k-CMlD, a constrained version of it which reflects Graph Motif more faithfully. We then give a fast algorithm for k-CMlD. As a result we obtain faster parameterized algorithms for Graph Motif and variants of it.