IEEE Information Theory Workshop (ITW)
Asymptotic analysis for hidden Markovian process with quantum hidden system
Masahito Hayashi
Abstract
We focus on a hidden Markovian process whose internal hidden system is given as a quantum system, and we address a sequence of data obtained from this process. Using a quantum version of Perron-Frobenius theorem, we derive novel upper and lower bounds for the cumulant generating function of the sample mean of the data. Using these bound, we derive the central limit theorem and large and moderate deviation for the tail probability. Further, our results can be extended to general probabilistic system.