Hamiltonicity of Sparse Pseudorandom Graphs

We show that every $(n,d,\lambda )$-graph contains a Hamilton cycle for sufficiently large $n$, assuming that $d\ge {\log }^{6}n$ and $\lambda \le cd$, where $c=\frac{1}{70000}$. This significantly improves a recent result of Glock, Correia and Sudakov, who obtained a similar result for $d$ that grows polynomially with $n$. The proof is based on a new result regarding the second largest eigenvalue of the adjacency matrix of a subgraph induced by a random subset of vertices, combined with a recent result on connecting designated pairs of vertices by vertex-disjoint paths in $(n,d,\lambda )$-graphs. We believe that the former result is of independent interest and will have further applications.