Why do explanations fail? A typology and discussion on failures in XAI
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_ | 1866909848626528256 |
|---|---|
| author | Bove, Clara Laugel, Thibault Lesot, Marie-Jeanne Tijus, Charles Detyniecki, Marcin |
| author_facet | Bove, Clara Laugel, Thibault Lesot, Marie-Jeanne Tijus, Charles Detyniecki, Marcin |
| contents | As Machine Learning models achieve unprecedented levels of performance, the XAI domain aims at making these models understandable by presenting end-users with intelligible explanations.
Yet, some existing XAI approaches fail to meet expectations: several issues have been reported in the literature, generally pointing out either
technical limitations or misinterpretations by users.
In this paper, we argue that the resulting harms arise from a complex overlap of multiple failures in XAI, which existing ad-hoc studies fail to capture.
This work therefore advocates for a holistic perspective, presenting a systematic investigation of limitations of current XAI methods and their impact on the interpretation of explanations. %
By distinguishing between system-specific and user-specific failures,
we propose a typological framework that helps revealing the nuanced complexities of explanation failures.
Leveraging this typology, we discuss some research directions to help practitioners better understand the limitations of XAI systems and enhance the quality of ML explanations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_13474 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Why do explanations fail? A typology and discussion on failures in XAI Bove, Clara Laugel, Thibault Lesot, Marie-Jeanne Tijus, Charles Detyniecki, Marcin Machine Learning Artificial Intelligence Human-Computer Interaction As Machine Learning models achieve unprecedented levels of performance, the XAI domain aims at making these models understandable by presenting end-users with intelligible explanations. Yet, some existing XAI approaches fail to meet expectations: several issues have been reported in the literature, generally pointing out either technical limitations or misinterpretations by users. In this paper, we argue that the resulting harms arise from a complex overlap of multiple failures in XAI, which existing ad-hoc studies fail to capture. This work therefore advocates for a holistic perspective, presenting a systematic investigation of limitations of current XAI methods and their impact on the interpretation of explanations. % By distinguishing between system-specific and user-specific failures, we propose a typological framework that helps revealing the nuanced complexities of explanation failures. Leveraging this typology, we discuss some research directions to help practitioners better understand the limitations of XAI systems and enhance the quality of ML explanations. |
| title | Why do explanations fail? A typology and discussion on failures in XAI |
| topic | Machine Learning Artificial Intelligence Human-Computer Interaction |
| url | https://arxiv.org/abs/2405.13474 |