Asymptotic optimality of finite model approximations for partially observed markov decision processes with discounted cost

Name: Asymptotic optimality of finite model approximations for partially observed markov decision processes with discounted cost
Author: Saldı, Naci, Yuksel, S., Linder, T.

İsim	Asymptotic optimality of finite model approximations for partially observed markov decision processes with discounted cost
Yazar	Saldı, Naci, Yuksel, S., Linder, T.
Basım Tarihi:	2020-01
Basım Yeri	- IEEE
Konu	Aerospace electronics, Convergence, Quantization (signal), Markov processes, Computational modeling, Cost function, Approximations, Markov decision processes, Non-linear filtering, Quantization, Stochastic control
Tür	Süreli Yayın
Dil	İngilizce
Dijital	Evet
Yazma	Hayır
Kütüphane:	Özyeğin Üniversitesi
Demirbaş Numarası	0018-9286
Kayıt Numarası	9f33c013-e783-4804-b33d-1fa2896875a9
Lokasyon	Natural and Mathematical Sciences
Tarih	2020-01
Notlar	Natural Sciences and Engineering Research Council of Canada (NSERC)
Örnek Metin	We consider finite model approximations of discrete-time partially observed Markov decision processes (POMDPs) under the discounted cost criterion. After converting the original partially observed stochastic control problem to a fully observed one on the belief space, the finite models are obtained through the uniform quantization of the state and action spaces of the belief space Markov decision process (MDP). Under mild assumptions on the components of the original model, it is established that the policies obtained from these finite models are nearly optimal for the belief space MDP, and so, for the original partially observed problem. The assumptions essentially require that the belief space MDP satisfies a mild weak continuity condition. We provide an example and introduce explicit approximation procedures for the quantization of the set of probability measures on the state space of POMDP (i.e., belief space).
DOI	10.1109/TAC.2019.2907172
Cilt	65

Kaynağa git Özyeğin Üniversitesi

Aramaya Dön

Özyeğin Üniversitesi

Kaynağa git

Asymptotic optimality of finite model approximations for partially observed markov decision processes with discounted cost

Yazar Saldı, Naci, Yuksel, S., Linder, T.

Basım Tarihi 2020-01

Basım Yeri - IEEE

Konu Aerospace electronics, Convergence, Quantization (signal), Markov processes, Computational modeling, Cost function, Approximations, Markov decision processes, Non-linear filtering, Quantization, Stochastic control

Tür Süreli Yayın

Dil İngilizce

Dijital Evet

Yazma Hayır

Kütüphane Özyeğin Üniversitesi

Demirbaş Numarası 0018-9286

Kayıt Numarası 9f33c013-e783-4804-b33d-1fa2896875a9

Lokasyon Natural and Mathematical Sciences

Tarih 2020-01

Notlar Natural Sciences and Engineering Research Council of Canada (NSERC)

Örnek Metin We consider finite model approximations of discrete-time partially observed Markov decision processes (POMDPs) under the discounted cost criterion. After converting the original partially observed stochastic control problem to a fully observed one on the belief space, the finite models are obtained through the uniform quantization of the state and action spaces of the belief space Markov decision process (MDP). Under mild assumptions on the components of the original model, it is established that the policies obtained from these finite models are nearly optimal for the belief space MDP, and so, for the original partially observed problem. The assumptions essentially require that the belief space MDP satisfies a mild weak continuity condition. We provide an example and introduce explicit approximation procedures for the quantization of the set of probability measures on the state space of POMDP (i.e., belief space).

DOI 10.1109/TAC.2019.2907172

Cilt 65