An Axiomatic Approach to Model-Agnostic Concept Explanations
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909072175923200 |
|---|---|
| author | Feng, Zhili Moshkovitz, Michal Di Castro, Dotan Kolter, J. Zico |
| author_facet | Feng, Zhili Moshkovitz, Michal Di Castro, Dotan Kolter, J. Zico |
| contents | Concept explanation is a popular approach for examining how human-interpretable concepts impact the predictions of a model. However, most existing methods for concept explanations are tailored to specific models. To address this issue, this paper focuses on model-agnostic measures. Specifically, we propose an approach to concept explanations that satisfy three natural axioms: linearity, recursivity, and similarity. We then establish connections with previous concept explanation methods, offering insight into their varying semantic meanings. Experimentally, we demonstrate the utility of the new method by applying it in different scenarios: for model selection, optimizer selection, and model improvement using a kind of prompt editing for zero-shot vision language models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_06890 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | An Axiomatic Approach to Model-Agnostic Concept Explanations Feng, Zhili Moshkovitz, Michal Di Castro, Dotan Kolter, J. Zico Machine Learning Concept explanation is a popular approach for examining how human-interpretable concepts impact the predictions of a model. However, most existing methods for concept explanations are tailored to specific models. To address this issue, this paper focuses on model-agnostic measures. Specifically, we propose an approach to concept explanations that satisfy three natural axioms: linearity, recursivity, and similarity. We then establish connections with previous concept explanation methods, offering insight into their varying semantic meanings. Experimentally, we demonstrate the utility of the new method by applying it in different scenarios: for model selection, optimizer selection, and model improvement using a kind of prompt editing for zero-shot vision language models. |
| title | An Axiomatic Approach to Model-Agnostic Concept Explanations |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2401.06890 |