ADVISE: ADaptive Feature Relevance and VISual Explanations for Convolutional Neural Networks
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2022
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910339579248640 |
|---|---|
| author | Dehshibi, Mohammad Mahdi Ashtari-Majlan, Mona Adhane, Gereziher Masip, David |
| author_facet | Dehshibi, Mohammad Mahdi Ashtari-Majlan, Mona Adhane, Gereziher Masip, David |
| contents | To equip Convolutional Neural Networks (CNNs) with explainability, it is essential to interpret how opaque models take specific decisions, understand what causes the errors, improve the architecture design, and identify unethical biases in the classifiers. This paper introduces ADVISE, a new explainability method that quantifies and leverages the relevance of each unit of the feature map to provide better visual explanations. To this end, we propose using adaptive bandwidth kernel density estimation to assign a relevance score to each unit of the feature map with respect to the predicted class. We also propose an evaluation protocol to quantitatively assess the visual explainability of CNN models. We extensively evaluate our idea in the image classification task using AlexNet, VGG16, ResNet50, and Xception pretrained on ImageNet. We compare ADVISE with the state-of-the-art visual explainable methods and show that the proposed method outperforms competing approaches in quantifying feature-relevance and visual explainability while maintaining competitive time complexity. Our experiments further show that ADVISE fulfils the sensitivity and implementation independence axioms while passing the sanity checks. The implementation is accessible for reproducibility purposes on https://github.com/dehshibi/ADVISE. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2203_01289 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | ADVISE: ADaptive Feature Relevance and VISual Explanations for Convolutional Neural Networks Dehshibi, Mohammad Mahdi Ashtari-Majlan, Mona Adhane, Gereziher Masip, David Computer Vision and Pattern Recognition Machine Learning To equip Convolutional Neural Networks (CNNs) with explainability, it is essential to interpret how opaque models take specific decisions, understand what causes the errors, improve the architecture design, and identify unethical biases in the classifiers. This paper introduces ADVISE, a new explainability method that quantifies and leverages the relevance of each unit of the feature map to provide better visual explanations. To this end, we propose using adaptive bandwidth kernel density estimation to assign a relevance score to each unit of the feature map with respect to the predicted class. We also propose an evaluation protocol to quantitatively assess the visual explainability of CNN models. We extensively evaluate our idea in the image classification task using AlexNet, VGG16, ResNet50, and Xception pretrained on ImageNet. We compare ADVISE with the state-of-the-art visual explainable methods and show that the proposed method outperforms competing approaches in quantifying feature-relevance and visual explainability while maintaining competitive time complexity. Our experiments further show that ADVISE fulfils the sensitivity and implementation independence axioms while passing the sanity checks. The implementation is accessible for reproducibility purposes on https://github.com/dehshibi/ADVISE. |
| title | ADVISE: ADaptive Feature Relevance and VISual Explanations for Convolutional Neural Networks |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2203.01289 |