Giant Components in Random Temporal Graphs

A temporal graph is a graph whose edges appear only at certain points in time. In these graphs, reachability among the nodes relies on paths that traverse edges in chronological order (temporal paths). Unlike standard paths, these paths are not always composable, thus the reachability relation is intransitive and connected components do not form equivalence classes. We investigate the properties of reachability and connected components in random temporal graphs, using a simple model that consists of permuting uniformly at random the edges of an Erd\"os-R\'enyi graph and interpreting the position in this permutation as presence times. This model was introduced in [Casteigts et al., FOCS 2021], where thresholds for various reachability properties were identified; for example, sharp threshold for temporal connectivity (all-to-all reachability) is $p=3 \log n / n$. We generalize several techniques from the above paper in order to characterize the emergence of a giant connected component, which answers an open question from that paper. The growth of a giant component turns out to be quite different from the static case, where a component of size $n^{2/3}$ emerges at $p_0=1/n$ and subsequently absorbs a constant fraction of all vertices, with this fraction gradually approaching 1. In contrast, in temporal graphs, we show that the size of a giant connected component transitions abruptly from $o(n)$ nodes to $n - o(n)$ nodes at $p = \log n / n$.