Evaluating Feature Attribution Methods in the Image Domain
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2022
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911982763900928 |
|---|---|
| author | Gevaert, Arne Rousseau, Axel-Jan Becker, Thijs Valkenborg, Dirk De Bie, Tijl Saeys, Yvan |
| author_facet | Gevaert, Arne Rousseau, Axel-Jan Becker, Thijs Valkenborg, Dirk De Bie, Tijl Saeys, Yvan |
| contents | Feature attribution maps are a popular approach to highlight the most important pixels in an image for a given prediction of a model. Despite a recent growth in popularity and available methods, little attention is given to the objective evaluation of such attribution maps. Building on previous work in this domain, we investigate existing metrics and propose new variants of metrics for the evaluation of attribution maps. We confirm a recent finding that different attribution metrics seem to measure different underlying concepts of attribution maps, and extend this finding to a larger selection of attribution metrics. We also find that metric results on one dataset do not necessarily generalize to other datasets, and methods with desirable theoretical properties such as DeepSHAP do not necessarily outperform computationally cheaper alternatives. Based on these findings, we propose a general benchmarking approach to identify the ideal feature attribution method for a given use case. Implementations of attribution metrics and our experiments are available online. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2202_12270 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Evaluating Feature Attribution Methods in the Image Domain Gevaert, Arne Rousseau, Axel-Jan Becker, Thijs Valkenborg, Dirk De Bie, Tijl Saeys, Yvan Computer Vision and Pattern Recognition Machine Learning Feature attribution maps are a popular approach to highlight the most important pixels in an image for a given prediction of a model. Despite a recent growth in popularity and available methods, little attention is given to the objective evaluation of such attribution maps. Building on previous work in this domain, we investigate existing metrics and propose new variants of metrics for the evaluation of attribution maps. We confirm a recent finding that different attribution metrics seem to measure different underlying concepts of attribution maps, and extend this finding to a larger selection of attribution metrics. We also find that metric results on one dataset do not necessarily generalize to other datasets, and methods with desirable theoretical properties such as DeepSHAP do not necessarily outperform computationally cheaper alternatives. Based on these findings, we propose a general benchmarking approach to identify the ideal feature attribution method for a given use case. Implementations of attribution metrics and our experiments are available online. |
| title | Evaluating Feature Attribution Methods in the Image Domain |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2202.12270 |