Mean field for Markov Decision Processes: from Discrete to Continuous Optimization - Archive ouverte HAL Access content directly
Reports Year :

Mean field for Markov Decision Processes: from Discrete to Continuous Optimization

Abstract

We study the convergence of Markov Decision Processes made of a large number of objects to optimization problems on ordinary differential equations (ODE). We show that the optimal reward of such a Markov Decision Process, satisfying a Bellman equation, converges to the solution of a continuous Hamilton-Jacobi-Bellman (HJB) equation based on the mean field approximation of the Markov Decision Process. We give bounds on the difference of the rewards, and a constructive algorithm for deriving an approximating solution to the Markov Decision Process from a solution of the HJB equations. We illustrate the method on three examples pertaining respectively to investment strategies, population dynamics control and scheduling in queues are developed. They are used to illustrate and justify the construction of the controlled ODE and to show the gain obtained by solving a continuous HJB equation rather than a large discrete Bellman equation.
Fichier principal
Vignette du fichier
RR_7239_MeanFieldMDP.pdf (383.24 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00473005 , version 1 (14-04-2010)
hal-00473005 , version 2 (02-07-2010)
hal-00473005 , version 3 (17-05-2011)

Identifiers

Cite

Nicolas Gast, Bruno Gaujal, Jean-Yves Le Boudec. Mean field for Markov Decision Processes: from Discrete to Continuous Optimization. 2010. ⟨hal-00473005v3⟩
521 View
580 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More