Hao, Meiling, Su, Pingfan, Hu, Liyuan, Szabo, Zoltan 
ORCID: 0000-0001-6183-7603, Zhao, Qianyu and Shi, Chengchun ORCID: 0000-0001-7773-2099
(2024)
Forward and backward state abstractions for off-policy evaluation.
.
arXiv.
(Submitted)
![[img]](http://eprints.lse.ac.uk/style/images/fileicons/text.png) |
Text (Szabo_forward-and-backward-state-abstractions)
- Submitted Version
Download (1MB) |
Abstract
Off-policy evaluation (OPE) is crucial for evaluating a target policy’s impact offline before its deployment. However, achieving accurate OPE in large state spaces remains challenging. This paper studies state abstractions – originally designed for policy learning – in the context of OPE. Our contributions are three-fold: (i) We define a set of irrelevance conditions central to learning state abstractions for OPE. (ii) We derive sufficient conditions for achieving irrelevance in Q-functions and marginalized importance sampling ratios, the latter obtained by constructing a time-reversed Markov decision process (MDP) based on the observed MDP. (iii) We propose a novel two-step procedure that sequentially projects the original state space into a smaller space, which substantially simplify the sample complexity of OPE arising from high cardinality.
| Item Type: | Monograph (Report) |
|---|---|
| Official URL: | https://arxiv.org/ |
| Divisions: | Statistics |
| Subjects: | H Social Sciences > HA Statistics |
| Date Deposited: | 02 Jul 2024 07:54 |
| Last Modified: | 25 Oct 2024 10:09 |
| URI: | http://eprints.lse.ac.uk/id/eprint/124074 |
Actions (login required)
 |
View Item |
Download Statistics
Download Statistics