Geometric Remove-and-Retrain (GOAR): Coordinate-Invariant eXplainable AI Assessment
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910531113189376 |
|---|---|
| author | Park, Yong-Hyun Seo, Junghoon Park, Bomseok Lee, Seongsu Jo, Junghyo |
| author_facet | Park, Yong-Hyun Seo, Junghoon Park, Bomseok Lee, Seongsu Jo, Junghyo |
| contents | Identifying the relevant input features that have a critical influence on the output results is indispensable for the development of explainable artificial intelligence (XAI). Remove-and-Retrain (ROAR) is a widely accepted approach for assessing the importance of individual pixels by measuring changes in accuracy following their removal and subsequent retraining of the modified dataset. However, we uncover notable limitations in pixel-perturbation strategies. When viewed from a geometric perspective, we discover that these metrics fail to discriminate between differences among feature attribution methods, thereby compromising the reliability of the evaluation. To address this challenge, we introduce an alternative feature-perturbation approach named Geometric Remove-and-Retrain (GOAR). Through a series of experiments with both synthetic and real datasets, we substantiate that GOAR transcends the limitations of pixel-centric metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_12401 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Geometric Remove-and-Retrain (GOAR): Coordinate-Invariant eXplainable AI Assessment Park, Yong-Hyun Seo, Junghoon Park, Bomseok Lee, Seongsu Jo, Junghyo Machine Learning Computer Vision and Pattern Recognition Identifying the relevant input features that have a critical influence on the output results is indispensable for the development of explainable artificial intelligence (XAI). Remove-and-Retrain (ROAR) is a widely accepted approach for assessing the importance of individual pixels by measuring changes in accuracy following their removal and subsequent retraining of the modified dataset. However, we uncover notable limitations in pixel-perturbation strategies. When viewed from a geometric perspective, we discover that these metrics fail to discriminate between differences among feature attribution methods, thereby compromising the reliability of the evaluation. To address this challenge, we introduce an alternative feature-perturbation approach named Geometric Remove-and-Retrain (GOAR). Through a series of experiments with both synthetic and real datasets, we substantiate that GOAR transcends the limitations of pixel-centric metrics. |
| title | Geometric Remove-and-Retrain (GOAR): Coordinate-Invariant eXplainable AI Assessment |
| topic | Machine Learning Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2407.12401 |