On Measuring Intrinsic Causal Attributions in Deep Neural Networks
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909723853324288 |
|---|---|
| author | Saha, Saptarshi Rathore, Dhruv Vansraj Saha, Soumadeep Garain, Utpal Doermann, David |
| author_facet | Saha, Saptarshi Rathore, Dhruv Vansraj Saha, Soumadeep Garain, Utpal Doermann, David |
| contents | 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 reliable explanations compared to existing global explanation techniques. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_09660 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | On Measuring Intrinsic Causal Attributions in Deep Neural Networks Saha, Saptarshi Rathore, Dhruv Vansraj Saha, Soumadeep Garain, Utpal Doermann, David Machine Learning 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 reliable explanations compared to existing global explanation techniques. |
| title | On Measuring Intrinsic Causal Attributions in Deep Neural Networks |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2505.09660 |