The Parameterised Complexity of Temporal Motif Counting, and a Lovász-Style Isomorphism Theorem

Abstract

We study the structural expressivity and the parameterised complexity of counting homomorphisms from small temporal patterns to large temporal graphs. Here, a temporal pattern $P$ consists of a graph together with a partial order on its edges, and a homomorphism from $P$ to a temporal graph must not only preserve edges, but also satisfy the temporal constraints imposed by the partial order of the edge set of the pattern. The main results of this work are three-fold: First, we prove a temporal Lovász-style theorem, stating that two temporal graphs are isomorphic (under a natural definition of temporal isomorphisms) if and only if they have the same number of homomorphisms from all temporal patterns. Second, we introduce a cliquewidth-based measure on temporal patterns, called the temporally order-augmented dual width, the "toadwidth" for short, and show that counting temporal homomorphisms is fixed-parameter tractable for temporal patterns of bounded toadwidth. Third, we provide a parameterised complexity dichotomy with an explicit tractability criterion for counting homomorphisms from totally ordered temporal patterns, classified along their underlying graph structure.