Clustering Context in Off-Policy Evaluation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912252701966336 |
|---|---|
| author | Guzman-Olivares, Daniel Schmidt, Philipp Golebiowski, Jacek Bekasov, Artur |
| author_facet | Guzman-Olivares, Daniel Schmidt, Philipp Golebiowski, Jacek Bekasov, Artur |
| contents | Off-policy evaluation can leverage logged data to estimate the effectiveness of new policies in e-commerce, search engines, media streaming services, or automatic diagnostic tools in healthcare. However, the performance of baseline off-policy estimators like IPS deteriorates when the logging policy significantly differs from the evaluation policy. Recent work proposes sharing information across similar actions to mitigate this problem. In this work, we propose an alternative estimator that shares information across similar contexts using clustering. We study the theoretical properties of the proposed estimator, characterizing its bias and variance under different conditions. We also compare the performance of the proposed estimator and existing approaches in various synthetic problems, as well as a real-world recommendation dataset. Our experimental results confirm that clustering contexts improves estimation accuracy, especially in deficient information settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_21304 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Clustering Context in Off-Policy Evaluation Guzman-Olivares, Daniel Schmidt, Philipp Golebiowski, Jacek Bekasov, Artur Machine Learning Artificial Intelligence Off-policy evaluation can leverage logged data to estimate the effectiveness of new policies in e-commerce, search engines, media streaming services, or automatic diagnostic tools in healthcare. However, the performance of baseline off-policy estimators like IPS deteriorates when the logging policy significantly differs from the evaluation policy. Recent work proposes sharing information across similar actions to mitigate this problem. In this work, we propose an alternative estimator that shares information across similar contexts using clustering. We study the theoretical properties of the proposed estimator, characterizing its bias and variance under different conditions. We also compare the performance of the proposed estimator and existing approaches in various synthetic problems, as well as a real-world recommendation dataset. Our experimental results confirm that clustering contexts improves estimation accuracy, especially in deficient information settings. |
| title | Clustering Context in Off-Policy Evaluation |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2502.21304 |