Answers are grounded in this paper and cite their sources.
For a simple random walk on a finite connected simple graph, we study how cut vertices create asymmetry of hitting times. Let c be a cut vertex with branches G_1, ..., G_s, and write m_i = |E(G_i)| and M = |E(G)|. Using the cut-vertex decomposition of hitting times, we derive, for every u in V(G_i) excluding c, the local-to-global ratio formula: H_G(c,u) / H_G(u,c) = (M - m_i) / m_i + (M / m_i) * [H_Gi(c,u) / H_Gi(u,c)]. In particular, H_G(c,u) / H_G(u,c) > (M - m_i) / m_i. Consequently, at least s - 1 branches G_i satisfy H_G(c,u) > H_G(u,c) for every u in V(G_i) excluding c. On a branch having the minimum number of edges, the ratio is greater than s - 1. We show that this constant is asymptotically best possible. These results imply that no finite connected simple graph with a cut vertex is highly symmetric and that K_2 is the only finite connected simple highly symmetric graph with a cut edge. This answers two questions posed by Palacios and Markowsky. We also combine the general reversibility identity for hitting times with the cut-vertex formulas to express the asymmetry between vertices lying in different branches explicitly in terms of local branch data.
No takes yet. Share an insight, caveat, or question.
Tamura et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: