Stable adiabatic times for Markov chains

From MaRDI portal
Publication:2804560




Abstract: In this paper we continue our work on adiabatic time of time-inhomogeneous Markov chains first introduced in Kovchegov (2010) and Bradford and Kovchegov (2011). Our study is an analog to the well-known Quantum Adiabatic (QA) theorem which characterizes the quantum adiabatic time for the evolution of a quantum system as a result of applying of a series of Hamilton operators, each is a linear combination of two given initial and final Hamilton operators, i.e. mathbfH(s)=(1s)mathbfH0+smathbfH1. Informally, the quantum adiabatic time of a quantum system specifies the speed at which the Hamiltonian operators changes so that the ground state of the system at any time s will always remain epsilon-close to that induced by the Hamilton operator mathbfH(s) at time s. Analogously, we derive a sufficient condition for the stable adiabatic time of a time-inhomogeneous Markov evolution specified by applying a series of transition probability matrices, each is a linear combination of two given irreducible and aperiodic transition probability matrices, i.e., mathbfPt=(1t)mathbfP0+tmathbfP1. In particular we show that the stable adiabatic time tsad(mathbfP0,mathbfP1,epsilon)=O(tmix4(epsilonslash2)slashepsilon3), where tmix denotes the maximum mixing time over all mathbfPt for 0leqtleq1.









This page was built for publication: Stable adiabatic times for Markov chains

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2804560)