Change-point detection for Piecewise Deterministic Markov Processes
DE SAPORTA, Benoîte
Quality control and dynamic reliability [CQFD]
Institut Montpelliérain Alexander Grothendieck [IMAG]
Quality control and dynamic reliability [CQFD]
Institut Montpelliérain Alexander Grothendieck [IMAG]
DE SAPORTA, Benoîte
Quality control and dynamic reliability [CQFD]
Institut Montpelliérain Alexander Grothendieck [IMAG]
< Leer menos
Quality control and dynamic reliability [CQFD]
Institut Montpelliérain Alexander Grothendieck [IMAG]
Idioma
en
Article de revue
Este ítem está publicado en
Automatica. 2018, vol. 97, p. 234-247
Elsevier
Resumen en inglés
We consider a change-point detection problem for a simple class of Piecewise Deterministic Markov Processes (PDMPs). A continuous-time PDMP is observed in discrete time and through noise, and the aim is to propose a numerical ...Leer más >
We consider a change-point detection problem for a simple class of Piecewise Deterministic Markov Processes (PDMPs). A continuous-time PDMP is observed in discrete time and through noise, and the aim is to propose a numerical method to accurately detect both the date of the change of dynamics and the new regime after the change. To do so, we state the problem as an optimal stopping problem for a partially observed discrete-time Markov decision process taking values in a continuous state space and provide a discretization of the state space based on quantization to approximate the value function and build a tractable stopping policy. We provide error bounds for the approximation of the value function and numerical simulations to assess the performance of our candidate policy.< Leer menos
Orígen
Importado de HalCentros de investigación