Spectral Representation for Causal Estimation with Hidden Confounders
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913729333952512 |
|---|---|
| author | Sun, Haotian Moulin, Antoine Ren, Tongzheng Gretton, Arthur Dai, Bo |
| author_facet | Sun, Haotian Moulin, Antoine Ren, Tongzheng Gretton, Arthur Dai, Bo |
| contents | We address the problem of causal effect estimation where hidden confounders are present, with a focus on two settings: instrumental variable regression with additional observed confounders, and proxy causal learning. Our approach uses a singular value decomposition of a conditional expectation operator, followed by a saddle-point optimization problem, which, in the context of IV regression, can be thought of as a neural net generalization of the seminal approach due to Darolles et al. [2011]. Saddle-point formulations have gathered considerable attention recently, as they can avoid double sampling bias and are amenable to modern function approximation methods. We provide experimental validation in various settings, and show that our approach outperforms existing methods on common benchmarks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_10448 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Spectral Representation for Causal Estimation with Hidden Confounders Sun, Haotian Moulin, Antoine Ren, Tongzheng Gretton, Arthur Dai, Bo Machine Learning We address the problem of causal effect estimation where hidden confounders are present, with a focus on two settings: instrumental variable regression with additional observed confounders, and proxy causal learning. Our approach uses a singular value decomposition of a conditional expectation operator, followed by a saddle-point optimization problem, which, in the context of IV regression, can be thought of as a neural net generalization of the seminal approach due to Darolles et al. [2011]. Saddle-point formulations have gathered considerable attention recently, as they can avoid double sampling bias and are amenable to modern function approximation methods. We provide experimental validation in various settings, and show that our approach outperforms existing methods on common benchmarks. |
| title | Spectral Representation for Causal Estimation with Hidden Confounders |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2407.10448 |