Functional Natural Policy Gradients
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917382994264064 |
|---|---|
| author | Bibaut, Aurelien Zenati, Houssam Rahier, Thibaud Kallus, Nathan |
| author_facet | Bibaut, Aurelien Zenati, Houssam Rahier, Thibaud Kallus, Nathan |
| contents | We propose a cross-fitted debiasing device for policy learning from offline data. A key consequence of the resulting learning principle is $\sqrt N$ regret even for policy classes with complexity greater than Donsker, provided a product-of-errors nuisance remainder is $O(N^{-1/2})$. The regret bound factors into a plug-in policy error factor governed by policy-class complexity and an environment nuisance factor governed by the complexity of the environment dynamics, making explicit how one may be traded against the other. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_28681 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Functional Natural Policy Gradients Bibaut, Aurelien Zenati, Houssam Rahier, Thibaud Kallus, Nathan Machine Learning We propose a cross-fitted debiasing device for policy learning from offline data. A key consequence of the resulting learning principle is $\sqrt N$ regret even for policy classes with complexity greater than Donsker, provided a product-of-errors nuisance remainder is $O(N^{-1/2})$. The regret bound factors into a plug-in policy error factor governed by policy-class complexity and an environment nuisance factor governed by the complexity of the environment dynamics, making explicit how one may be traded against the other. |
| title | Functional Natural Policy Gradients |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2603.28681 |