Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2406.13257 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918284621774848 |
|---|---|
| author | Rodrigues, Caroline Mazini Boutry, Nicolas Najman, Laurent |
| author_facet | Rodrigues, Caroline Mazini Boutry, Nicolas Najman, Laurent |
| contents | Challenges persist in providing interpretable explanations for neural network reasoning in explainable AI (xAI). Existing methods like Integrated Gradients produce noisy maps, and LIME, while intuitive, may deviate from the model's reasoning. We introduce a framework that uses hierarchical segmentation techniques for faithful and interpretable explanations of Convolutional Neural Networks (CNNs). Our method constructs model-based hierarchical segmentations that maintain the model's reasoning fidelity and allows both human-centric and model-centric segmentation. This approach offers multiscale explanations, aiding bias identification and enhancing understanding of neural network decision-making. Experiments show that our framework, xAiTrees, delivers highly interpretable and faithful model explanations, not only surpassing traditional xAI methods but shedding new light on a novel approach to enhancing xAI interpretability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_13257 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Explaning with trees: interpreting CNNs using hierarchies Rodrigues, Caroline Mazini Boutry, Nicolas Najman, Laurent Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning Challenges persist in providing interpretable explanations for neural network reasoning in explainable AI (xAI). Existing methods like Integrated Gradients produce noisy maps, and LIME, while intuitive, may deviate from the model's reasoning. We introduce a framework that uses hierarchical segmentation techniques for faithful and interpretable explanations of Convolutional Neural Networks (CNNs). Our method constructs model-based hierarchical segmentations that maintain the model's reasoning fidelity and allows both human-centric and model-centric segmentation. This approach offers multiscale explanations, aiding bias identification and enhancing understanding of neural network decision-making. Experiments show that our framework, xAiTrees, delivers highly interpretable and faithful model explanations, not only surpassing traditional xAI methods but shedding new light on a novel approach to enhancing xAI interpretability. |
| title | Explaning with trees: interpreting CNNs using hierarchies |
| topic | Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2406.13257 |