Journal of the Korean Operations Research and Management Science Society (한국경영과학회지)

The Korean Operations Research and Management Science Society (한국경영과학회)
권호 표지

Partially Observable Markov Decision Process with Lagged Information over Infinite Horizon

  • Published : 1991.06.01
Download PDF
⟨ Previous Next ⟩

Abstract

This paper shows the infinite horizon model of Partially Observable Markov Decision Process with lagged information. The lagged information is uncertain delayed observation of the process under control. Even though the optimal policy of the model exists, finding the optimal policy is very time consuming. Thus, the aim of this study is to find an .eplison.-optimal stationary policy minimizing the expected discounted total cost of the model. .EPSILON.- optimal policy is found by using a modified version of the well known policy iteration algorithm. The modification focuses to the value determination routine of the algorithm. Some properties of the approximation functions for the expected discounted cost of a stationary policy are presented. The expected discounted cost of a stationary policy is approximated based on these properties. A numerical example is also shown.

Keywords

References

  1. Oper. Res. v.27 Structural Results for Partially Observable Markov Decision Processes Albright,S.C.
  2. Ann. Math. Stat. v.36 Discounted Dynamic Programming Blackwell,D.
  3. SIAM Rev. v.9 Contraction Mappings in the Theory Underlying Dynamic Programming Denardo,E.V.
  4. J. Opl. Res. Soc. v.38 A Partially Obsevable Markov Decision Process with Lagged Information Kim,S.H.;Jeong,B.H.
  5. Unpublished Dissertation, Korea Advanced Institute of Science and Technology Use of Lagged Information in partially Observable Markov Decision Process Jeong,B.H.
  6. SIAM J. Con. & Opt. v.18 Optimal Infinite-Horizon Undiscounted Control of Finite Probabilistic Systems Platzman,L.K.
  7. Recent Developments in Markov Decision Prcoesses Piecewise Linear Markov Decision Process with an aplication into Patially Observable Models Sawaki,K.;R.Hartley(et al.)(ed.)
  8. J. Math. Anal. & Appl. v.91 Transformation of Partially Observable Markov Decision Processes into Piecewise Linear Ones Sawaki,K.
  9. Oper. Res. v.26 The Optimal Control of Partially Observable Markov Process over the Infinite Horizon; Discounted Costs Sondik,E.J.
  10. Euro. J. Oper. Res. v.21 A Markov Decision Algorithm for Optmal Inspections and Revisions in a Maintenance System with Partial Information Tijms,H.C.;F.A.van der Duyn Schouten
  11. J. Opl. Res. Sov. v.39 Note on 'A Partially Observable Markov Decision Process with Lagged Information' White,C.C.
  12. J. Math. Anal. & Appl. v.72 Finite State Approximations for Denumerable State Infinite Horizon Contractd Markov Decision Processes: The Policy Space Method White,D.J.