Fast Explanations via Policy Gradient-Optimized Explainer
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914376659763200 |
|---|---|
| author | Pan, Deng Moniz, Nuno Chawla, Nitesh |
| author_facet | Pan, Deng Moniz, Nuno Chawla, Nitesh |
| contents | The challenge of delivering efficient explanations is a critical barrier that prevents the adoption of model explanations in real-world applications. Existing approaches often depend on extensive model queries for sample-level explanations or rely on expert's knowledge of specific model structures that trade general applicability for efficiency. To address these limitations, this paper introduces a novel framework Fast Explanation (FEX) that represents attribution-based explanations via probability distributions, which are optimized by leveraging the policy gradient method. The proposed framework offers a robust, scalable solution for real-time, large-scale model explanations, bridging the gap between efficiency and applicability. We validate our framework on image and text classification tasks and the experiments demonstrate that our method reduces inference time by over 97% and memory usage by 70% compared to traditional model-agnostic approaches while maintaining high-quality explanations and broad applicability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_18664 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Fast Explanations via Policy Gradient-Optimized Explainer Pan, Deng Moniz, Nuno Chawla, Nitesh Machine Learning Artificial Intelligence The challenge of delivering efficient explanations is a critical barrier that prevents the adoption of model explanations in real-world applications. Existing approaches often depend on extensive model queries for sample-level explanations or rely on expert's knowledge of specific model structures that trade general applicability for efficiency. To address these limitations, this paper introduces a novel framework Fast Explanation (FEX) that represents attribution-based explanations via probability distributions, which are optimized by leveraging the policy gradient method. The proposed framework offers a robust, scalable solution for real-time, large-scale model explanations, bridging the gap between efficiency and applicability. We validate our framework on image and text classification tasks and the experiments demonstrate that our method reduces inference time by over 97% and memory usage by 70% compared to traditional model-agnostic approaches while maintaining high-quality explanations and broad applicability. |
| title | Fast Explanations via Policy Gradient-Optimized Explainer |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2405.18664 |