Cargando…
To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods
The main objective of eXplainable Artificial Intelligence (XAI) is to provide effective explanations for black-box classifiers. The existing literature lists many desirable properties for explanations to be useful, but there is a scarce consensus on how to quantitatively evaluate explanations in pra...
Autores principales: | , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
PeerJ Inc.
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8056245/ https://www.ncbi.nlm.nih.gov/pubmed/33977131 http://dx.doi.org/10.7717/peerj-cs.479 |
_version_ | 1783680613148000256 |
---|---|
author | Amparore, Elvio Perotti, Alan Bajardi, Paolo |
author_facet | Amparore, Elvio Perotti, Alan Bajardi, Paolo |
author_sort | Amparore, Elvio |
collection | PubMed |
description | The main objective of eXplainable Artificial Intelligence (XAI) is to provide effective explanations for black-box classifiers. The existing literature lists many desirable properties for explanations to be useful, but there is a scarce consensus on how to quantitatively evaluate explanations in practice. Moreover, explanations are typically used only to inspect black-box models, and the proactive use of explanations as a decision support is generally overlooked. Among the many approaches to XAI, a widely adopted paradigm is Local Linear Explanations—with LIME and SHAP emerging as state-of-the-art methods. We show that these methods are plagued by many defects including unstable explanations, divergence of actual implementations from the promised theoretical properties, and explanations for the wrong label. This highlights the need to have standard and unbiased evaluation procedures for Local Linear Explanations in the XAI field. In this paper we address the problem of identifying a clear and unambiguous set of metrics for the evaluation of Local Linear Explanations. This set includes both existing and novel metrics defined specifically for this class of explanations. All metrics have been included in an open Python framework, named LEAF. The purpose of LEAF is to provide a reference for end users to evaluate explanations in a standardised and unbiased way, and to guide researchers towards developing improved explainable techniques. |
format | Online Article Text |
id | pubmed-8056245 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2021 |
publisher | PeerJ Inc. |
record_format | MEDLINE/PubMed |
spelling | pubmed-80562452021-05-10 To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods Amparore, Elvio Perotti, Alan Bajardi, Paolo PeerJ Comput Sci Artificial Intelligence The main objective of eXplainable Artificial Intelligence (XAI) is to provide effective explanations for black-box classifiers. The existing literature lists many desirable properties for explanations to be useful, but there is a scarce consensus on how to quantitatively evaluate explanations in practice. Moreover, explanations are typically used only to inspect black-box models, and the proactive use of explanations as a decision support is generally overlooked. Among the many approaches to XAI, a widely adopted paradigm is Local Linear Explanations—with LIME and SHAP emerging as state-of-the-art methods. We show that these methods are plagued by many defects including unstable explanations, divergence of actual implementations from the promised theoretical properties, and explanations for the wrong label. This highlights the need to have standard and unbiased evaluation procedures for Local Linear Explanations in the XAI field. In this paper we address the problem of identifying a clear and unambiguous set of metrics for the evaluation of Local Linear Explanations. This set includes both existing and novel metrics defined specifically for this class of explanations. All metrics have been included in an open Python framework, named LEAF. The purpose of LEAF is to provide a reference for end users to evaluate explanations in a standardised and unbiased way, and to guide researchers towards developing improved explainable techniques. PeerJ Inc. 2021-04-16 /pmc/articles/PMC8056245/ /pubmed/33977131 http://dx.doi.org/10.7717/peerj-cs.479 Text en ©2021 Amparore et al. https://creativecommons.org/licenses/by/4.0/This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Computer Science) and either DOI or URL of the article must be cited. |
spellingShingle | Artificial Intelligence Amparore, Elvio Perotti, Alan Bajardi, Paolo To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods |
title | To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods |
title_full | To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods |
title_fullStr | To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods |
title_full_unstemmed | To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods |
title_short | To trust or not to trust an explanation: using LEAF to evaluate local linear XAI methods |
title_sort | to trust or not to trust an explanation: using leaf to evaluate local linear xai methods |
topic | Artificial Intelligence |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8056245/ https://www.ncbi.nlm.nih.gov/pubmed/33977131 http://dx.doi.org/10.7717/peerj-cs.479 |
work_keys_str_mv | AT amparoreelvio totrustornottotrustanexplanationusingleaftoevaluatelocallinearxaimethods AT perottialan totrustornottotrustanexplanationusingleaftoevaluatelocallinearxaimethods AT bajardipaolo totrustornottotrustanexplanationusingleaftoevaluatelocallinearxaimethods |