Two-way Deconfounder for Off-policy Evaluation in Causal Reinforcement Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910733390839808 |
|---|---|
| author | Yu, Shuguang Fang, Shuxing Peng, Ruixin Qi, Zhengling Zhou, Fan Shi, Chengchun |
| author_facet | Yu, Shuguang Fang, Shuxing Peng, Ruixin Qi, Zhengling Zhou, Fan Shi, Chengchun |
| contents | This paper studies off-policy evaluation (OPE) in the presence of unmeasured confounders. Inspired by the two-way fixed effects regression model widely used in the panel data literature, we propose a two-way unmeasured confounding assumption to model the system dynamics in causal reinforcement learning and develop a two-way deconfounder algorithm that devises a neural tensor network to simultaneously learn both the unmeasured confounders and the system dynamics, based on which a model-based estimator can be constructed for consistent policy value estimation. We illustrate the effectiveness of the proposed estimator through theoretical results and numerical experiments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_05783 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Two-way Deconfounder for Off-policy Evaluation in Causal Reinforcement Learning Yu, Shuguang Fang, Shuxing Peng, Ruixin Qi, Zhengling Zhou, Fan Shi, Chengchun Machine Learning This paper studies off-policy evaluation (OPE) in the presence of unmeasured confounders. Inspired by the two-way fixed effects regression model widely used in the panel data literature, we propose a two-way unmeasured confounding assumption to model the system dynamics in causal reinforcement learning and develop a two-way deconfounder algorithm that devises a neural tensor network to simultaneously learn both the unmeasured confounders and the system dynamics, based on which a model-based estimator can be constructed for consistent policy value estimation. We illustrate the effectiveness of the proposed estimator through theoretical results and numerical experiments. |
| title | Two-way Deconfounder for Off-policy Evaluation in Causal Reinforcement Learning |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2412.05783 |