On Gradient-like Explanation under a Black-box Setting: When Black-box Explanations Become as Good as White-box
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866916244694761472 |
|---|---|
| author | Cai, Yi Wunder, Gerhard |
| author_facet | Cai, Yi Wunder, Gerhard |
| contents | Attribution methods shed light on the explainability of data-driven approaches such as deep learning models by uncovering the most influential features in a to-be-explained decision. While determining feature attributions via gradients delivers promising results, the internal access required for acquiring gradients can be impractical under safety concerns, thus limiting the applicability of gradient-based approaches. In response to such limited flexibility, this paper presents \methodAbr~(gradient-estimation-based explanation), an approach that produces gradient-like explanations through only query-level access. The proposed approach holds a set of fundamental properties for attribution methods, which are mathematically rigorously proved, ensuring the quality of its explanations. In addition to the theoretical analysis, with a focus on image data, the experimental results empirically demonstrate the superiority of the proposed method over state-of-the-art black-box methods and its competitive performance compared to methods with full access. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2308_09381 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | On Gradient-like Explanation under a Black-box Setting: When Black-box Explanations Become as Good as White-box Cai, Yi Wunder, Gerhard Machine Learning Attribution methods shed light on the explainability of data-driven approaches such as deep learning models by uncovering the most influential features in a to-be-explained decision. While determining feature attributions via gradients delivers promising results, the internal access required for acquiring gradients can be impractical under safety concerns, thus limiting the applicability of gradient-based approaches. In response to such limited flexibility, this paper presents \methodAbr~(gradient-estimation-based explanation), an approach that produces gradient-like explanations through only query-level access. The proposed approach holds a set of fundamental properties for attribution methods, which are mathematically rigorously proved, ensuring the quality of its explanations. In addition to the theoretical analysis, with a focus on image data, the experimental results empirically demonstrate the superiority of the proposed method over state-of-the-art black-box methods and its competitive performance compared to methods with full access. |
| title | On Gradient-like Explanation under a Black-box Setting: When Black-box Explanations Become as Good as White-box |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2308.09381 |