Newest papers

Near Optimal Finite Model POMDP Approximation of Controlled Diffusion with Partial Information

Yunus Emre Demirci, Somnath Pradhan, Serdar Yüksel · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28377 · cited by 0 works (OpenAlex) · open-access PDF

Korean

우리의 주요 기여는 다음과 같습니다: (i) 우리는 근접한 최적화, 유한한 지평선 및 할인 된 무한한 지평선 비용 기준을 보여줍니다; (ii) 우리는 이러한 POMDP 대략 모델의 몇 가지 정규성 속성을 얻습니다; 그리고 (iii) 이러한 정규성 속성을 사용하여 우리는 최적화 된 최적화 된 핀 모델 대략에 도달합니다. (iv) 이들은 우리의 주요 결과로 고정된 최적화된 솔루션은 최적화 된 최적화 된 최소한의 POMDP 모델에 대한 최적화된 솔루션이 부분적인 정보로의 원래 확산에 가까운 최적화입니다.

Machine translation (Meta M2M100); the English original follows.

Abstract

We study continuous-time stochastic control problems under partial information. Our main contributions are the following: (i) We show near optimality, under finite horizon as well as discounted infinite horizon cost criteria, of piece-wise constant control policies obtained via solutions to discrete-time partially observable Markov decision process (POMDP) approximations, constructed via Euler-Maruyama approximations of both state and observation processes; (ii) we obtain several regularity properties of such POMDP approximate models; and (iii) using these regularity properties, we arrive at near optimal finite model approximations. (iv) These culminate into our main result which is that an optimal solution for the approximate finite POMDP model is near optimal for the original diffusion with partial information.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Partially observable Markov decision process, Optimal control, Mathematics, Observable, Markov process, Applied mathematics

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange