On Measuring Intrinsic Causal Attributions in Deep Neural Networks
Abstract
Quantifying the causal influence of input features within neural networks has become a topic of increasing interest. Existing approaches typically assess direct, indirect, and total causal effects. This work treats NNs as structural causal models (SCMs) and extends our focus to include intrinsic causal contributions (ICC). We propose an identifiable generative post-hoc framework for quantifying ICC. We also draw a relationship between ICC and Sobol’ indices. Our experiments on synthetic and real-world datasets demonstrate that ICC generates more intuitive and faithful explanations compared to existing global explanation techniques.
Cite
Text
Saha et al. "On Measuring Intrinsic Causal Attributions in Deep Neural Networks." Proceedings of the Fourth Conference on Causal Learning and Reasoning, 2025.Markdown
[Saha et al. "On Measuring Intrinsic Causal Attributions in Deep Neural Networks." Proceedings of the Fourth Conference on Causal Learning and Reasoning, 2025.](https://mlanthology.org/clear/2025/saha2025clear-measuring/)BibTeX
@inproceedings{saha2025clear-measuring,
title = {{On Measuring Intrinsic Causal Attributions in Deep Neural Networks}},
author = {Saha, Saptarshi and Rathore, Dhruv Vansraj and Saha, Soumadeep and Doermann, David and Garain, Utpal},
booktitle = {Proceedings of the Fourth Conference on Causal Learning and Reasoning},
year = {2025},
pages = {1405-1434},
volume = {275},
url = {https://mlanthology.org/clear/2025/saha2025clear-measuring/}
}