How slow, or fast, are standard random walks? - Analysis of hitting and cover times on trees
Yoshiaki Nonaka, Hirotaka Ono, Shuji Kijima, Masafumi Yamashita · Computing: The Australasian Theory Symposium · 2011
Random walk is a powerful tool, not only for modeling, but also for practical use such as the Internet crawlers. Standard random walks on graphs have been well studied; It is well-known that both hitting time and cover time of a standard random walk are bounded by O(n3) for any graph with n vertices, besides the bound is tight for some graphs. Ikeda et al. (2003) provided β-random walk, which realizes O(n2) hitting time and O(n2 log n) cover times for any graph, thus it archives, in a sense, n-times improvement compared to the standard random walk. This paper is concerned with optimizations of hitting and cover times, by drawing a comparison between the standard random walk and the fastest random walk. We show for any tree that the hitting time of the standard random walk is at most O(n)-times longer than one of the fastest random walk. Similarly, the cover time of the standard random walk is at most O(√n log n)-times longer than the fastest one, for any tree. We also show that our bound for the hitting time is tight by giving examples, while we only give a lower bound Ω(√n/log n) for the cover time.