Deep Backtracking Counterfactuals for Causally Compliant Explanations
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2023
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909283219668992 |
|---|---|
| author | Kladny, Klaus-Rudolf von Kügelgen, Julius Schölkopf, Bernhard Muehlebach, Michael |
| author_facet | Kladny, Klaus-Rudolf von Kügelgen, Julius Schölkopf, Bernhard Muehlebach, Michael |
| contents | Counterfactuals answer questions of what would have been observed under altered circumstances and can therefore offer valuable insights. Whereas the classical interventional interpretation of counterfactuals has been studied extensively, backtracking constitutes a less studied alternative where all causal laws are kept intact. In the present work, we introduce a practical method called deep backtracking counterfactuals (DeepBC) for computing backtracking counterfactuals in structural causal models that consist of deep generative components. We propose two distinct versions of our method--one utilizing Langevin Monte Carlo sampling and the other employing constrained optimization--to generate counterfactuals for high-dimensional data. As a special case, our formulation reduces to methods in the field of counterfactual explanations. Compared to these, our approach represents a causally compliant, versatile and modular alternative. We demonstrate these properties experimentally on a modified version of MNIST and CelebA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2310_07665 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Deep Backtracking Counterfactuals for Causally Compliant Explanations Kladny, Klaus-Rudolf von Kügelgen, Julius Schölkopf, Bernhard Muehlebach, Michael Artificial Intelligence Machine Learning Counterfactuals answer questions of what would have been observed under altered circumstances and can therefore offer valuable insights. Whereas the classical interventional interpretation of counterfactuals has been studied extensively, backtracking constitutes a less studied alternative where all causal laws are kept intact. In the present work, we introduce a practical method called deep backtracking counterfactuals (DeepBC) for computing backtracking counterfactuals in structural causal models that consist of deep generative components. We propose two distinct versions of our method--one utilizing Langevin Monte Carlo sampling and the other employing constrained optimization--to generate counterfactuals for high-dimensional data. As a special case, our formulation reduces to methods in the field of counterfactual explanations. Compared to these, our approach represents a causally compliant, versatile and modular alternative. We demonstrate these properties experimentally on a modified version of MNIST and CelebA. |
| title | Deep Backtracking Counterfactuals for Causally Compliant Explanations |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2310.07665 |