Robust hamiltonicity of random directed graphs

TL;DR

This paper proves that random directed graphs with sufficiently high edge probability are highly resilient in maintaining Hamiltonian cycles even after adversarial edge removals, extending classical results to sparser random settings.

Contribution

The paper improves the known bounds on the edge probability threshold for the local resilience of random directed graphs with respect to Hamiltonicity, approaching optimality up to polylogarithmic factors.

Findings

Established that the local resilience is close to 1/2 for p=ω(log^8 n / n)

Extended previous results to sparser random directed graphs

Achieved near-optimal bounds on edge probability for Hamiltonian resilience

Abstract

In his seminal paper from 1952 Dirac showed that the complete graph on $n\geq 3$ vertices remains Hamiltonian even if we allow an adversary to remove $\lfloor n/2\rfloor$ edges touching each vertex. In 1960 Ghouila-Houri obtained an analogue statement for digraphs by showing that every directed graph on $n\geq 3$ vertices with minimum in- and out-degree at least $n/2$ contains a directed Hamilton cycle. Both statements quantify the robustness of complete graphs (digraphs) with respect to the property of containing a Hamilton cycle. A natural way to generalize such results to arbitrary graphs (digraphs) is using the notion of \emph{local resilience}. The local resilience of a graph (digraph) $G$ with respect to a property $\mathcal{P}$ is the maximum number $r$ such that $G$ has the property $\mathcal{P}$ even if we allow an adversary to remove an $r$-fraction of (in- and out-going) edges touching each vertex. The theorems of Dirac and Ghouila-Houri state that the local resilience of the complete graph and digraph with respect to Hamiltonicity is $1/2$. Recently, this statements have been generalized to random settings. Lee and Sudakov (2012) proved that the local resilience of a random graph with edge probability $p=\omega(\log n /n)$ with respect to Hamiltonicity is $1/2\pm o(1)$. For random directed graphs, Hefetz, Steger and Sudakov (2014+) proved an analogue statement, but only for edge probability $p=\omega(\log n/\sqrt{n})$. In this paper we significantly improve their result to $p=\omega(\log^8 n/ n)$, which is optimal up to the polylogarithmic factor.