Neural Networks with Causal Graph Constraints: A New Approach for Treatment Effects Estimation
Fuente:
arXiv
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913320916746240 |
|---|---|
| author | Pros, Roger Vitrià, Jordi |
| author_facet | Pros, Roger Vitrià, Jordi |
| contents | In recent years, there has been a growing interest in using machine learning techniques for the estimation of treatment effects. Most of the best-performing methods rely on representation learning strategies that encourage shared behavior among potential outcomes to increase the precision of treatment effect estimates. In this paper we discuss and classify these models in terms of their algorithmic inductive biases and present a new model, NN-CGC, that considers additional information from the causal graph. NN-CGC tackles bias resulting from spurious variable interactions by implementing novel constraints on models, and it can be integrated with other representation learning methods. We test the effectiveness of our method using three different base models on common benchmarks. Our results indicate that our model constraints lead to significant improvements, achieving new state-of-the-art results in treatment effects estimation. We also show that our method is robust to imperfect causal graphs and that using partial causal information is preferable to ignoring it. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_12238 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Neural Networks with Causal Graph Constraints: A New Approach for Treatment Effects Estimation Pros, Roger Vitrià, Jordi Machine Learning Methodology In recent years, there has been a growing interest in using machine learning techniques for the estimation of treatment effects. Most of the best-performing methods rely on representation learning strategies that encourage shared behavior among potential outcomes to increase the precision of treatment effect estimates. In this paper we discuss and classify these models in terms of their algorithmic inductive biases and present a new model, NN-CGC, that considers additional information from the causal graph. NN-CGC tackles bias resulting from spurious variable interactions by implementing novel constraints on models, and it can be integrated with other representation learning methods. We test the effectiveness of our method using three different base models on common benchmarks. Our results indicate that our model constraints lead to significant improvements, achieving new state-of-the-art results in treatment effects estimation. We also show that our method is robust to imperfect causal graphs and that using partial causal information is preferable to ignoring it. |
| title | Neural Networks with Causal Graph Constraints: A New Approach for Treatment Effects Estimation |
| topic | Machine Learning Methodology |
| url | https://arxiv.org/abs/2404.12238 |