Menger's Theorem for Temporal Paths (Not Walks)

Abstract: A (directed) temporal graph is a (directed) graph whose edges are available only at specific times during its lifetime $\tau$. Temporal walks are sequences of adjacent edges whose appearing times are either strictly increasing or non-decreasing (here called non-strict), depending on the scenario. Paths are temporal walks where no vertex repetition is allowed. A temporal vertex is a pair $(u,i)$ where $u$ is a vertex and $i\in[\tau]$ a timestep. In this paper we focus on the questions: (i) are there at least $k$ paths from a single source $s$ to a single target $t$, no two of which internally intersect on a temporal vertex? (ii) are there at most $h$ temporal vertices whose removal disconnects $s$ from $t$? Let $k^*$ be the maximum value $k$ for which the answer to (i) is YES, and let $h^*$ be the minimum value $h$ for which the answer to (ii) is YES. In static graphs, $k^*$ and $h^*$ are equal by Menger's Theorem and this is a crucial property to solve efficiently both (i) and (ii). In temporal graphs such equality has been investigated only focusing on disjoint walks rather than disjoint paths. In this context, we prove that $k^*$ is equal to $h^*$ if and only if $k^*$ is 1. We show that this implies a dichotomy for (i), which turns out to be polynomial-time solvable when $k \le 2$, and NP-complete for $k \ge 3$. Finally, we give hardness results and an XP algorithm for (ii).