paper

Characterization of cutoff for reversible Markov chains

arXiv:1409.3250 · doi:10.1214/16-AOP1090

Abstract

A sequence of Markov chains is said to exhibit (total variation) cutoff if the convergence to stationarity in total variation distance is abrupt. We consider reversible lazy chains. We prove a necessary and sufficient condition for the occurrence of the cutoff phenomena in terms of concentration of hitting time of "worst" (in some sense) sets of stationary measure at least , for some . We also give general bounds on the total variation distance of a reversible chain at time in terms of the probability that some "worst" set of stationary measure at least was not hit by time . As an application of our techniques we show that a sequence of lazy Markov chains on finite trees exhibits a cutoff iff the ratio of their relaxation-times and their (lazy) mixing-times tends to 0.

Improved Theorem 3. Extended abstract appeared in SODA 2015

References in corpus (4)

Cited by in corpus (13)