𝔖 Bobbio Scriptorium
✦   LIBER   ✦

Empirical estimation in average Markov control processes

✍ Scribed by J. Adolfo Minjárez-Sosa

Publisher
Elsevier Science
Year
2008
Tongue
English
Weight
234 KB
Volume
21
Category
Article
ISSN
0893-9659

No coin nor oath required. For personal study only.

✦ Synopsis

This work concerns discrete-time Markov control processes with unbounded costs and unknown disturbance distribution θ. Assuming observability of the random disturbance, we estimate θ using its empirical estimator, which, combined with a variant of the vanishing discount factor approach, yields average cost optimal policies.

📜 SIMILAR VOLUMES

A pause control approach to the value it
A pause control approach to the value iteration scheme in average Markov decision processes
✍ Rolando Cavazos-Cadena 📂 Article 📅 1998 🏛 Elsevier Science 🌐 English ⚖ 124 KB

This work concerns average Markov decision chains with denumerable state space. Assuming that the Lyapunov function condition holds, it is shown that the value iteration scheme yields convergent approximations to the solution of the average cost optimality equation. This result is obtained using a p

Bayesian adaptive control of discrete-ti
Bayesian adaptive control of discrete-time Markov processes with long-run average cost
✍ G.B.Di Masi; Ł. Stettner 📂 Article 📅 1998 🏛 Elsevier Science 🌐 English ⚖ 100 KB

A simple adaptive control strategy for discrete-time Markov processes with compact state, action and parameter spaces that guarantees near self-optimality is proposed. The approach used is based on randomization and the study of invariant measure of the joint state and Bayesian parameter estimator p