Rida, Adam; Lesot, Marie-Jeanne; Renard, Xavier; Marsala, Christophe
Dynamic Interpretability for Model Comparison via Decision Rules Conference
2023 ECML PKDD Workshop on Explainable Artificial Intelligence From Static to Dynamic (DynXAI), 2023.
BibTeX | Tags: Explainability & Interpretability
@conference{nokey,
title = {Dynamic Interpretability for Model Comparison via Decision Rules},
author = {Adam Rida and Marie-Jeanne Lesot and Xavier Renard and Christophe Marsala},
year = {2023},
date = {2023-09-18},
urldate = {2023-09-18},
booktitle = {2023 ECML PKDD Workshop on Explainable Artificial Intelligence From Static to Dynamic (DynXAI)},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {conference}
}
Thibault Laugel Adulam Jeyasothy, Marie-Jeanne Lesot
A general framework for personalising post hoc explanations through user knowledge integration Journal Article
In: International Journal of Approximate Reasoning, vol. 160, 2023.
Links | BibTeX | Tags: Explainability & Interpretability
@article{jeyasothy2023,
title = {A general framework for personalising post hoc explanations through user knowledge integration},
author = {Adulam Jeyasothy, Thibault Laugel, Marie-Jeanne Lesot, Christophe Marsala, Marcin Detyniecki},
doi = {https://doi.org/10.1016/j.ijar.2023.108944},
year = {2023},
date = {2023-09-01},
journal = {International Journal of Approximate Reasoning},
volume = {160},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {article}
}
Adulam Jeyasothy Thibault Laugel, Marie-Jeanne Lesot
Achieving Diversity in Counterfactual Explanations: a Review and Discussion Conference
Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency, 2023.
Links | BibTeX | Tags: Explainability & Interpretability
@conference{laugel2023,
title = {Achieving Diversity in Counterfactual Explanations: a Review and Discussion},
author = {Thibault Laugel, Adulam Jeyasothy, Marie-Jeanne Lesot, Christophe Marsala, Marcin Detyniecki},
url = {https://arxiv.org/abs/2305.05840},
doi = {https://doi.org/10.1145/3593013.3594122},
year = {2023},
date = {2023-06-15},
booktitle = {Proceedings of the 2023 ACM Conference on Fairness, Accountability, and Transparency},
pages = {1859--1869},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {conference}
}
Bove, Clara; Lesot, Marie-Jeanne; Tijus, Charles; Detyniecki, Marcin
Investigating the Intelligibility of Plural Counterfactual Examples for Non-Expert Users, an Explanation User Interface Proposition and User Study Conference
Proceedings of the 28th International Conference on Intelligent User Interfaces, IUI'23, 2023.
BibTeX | Tags: Explainability & Interpretability
@conference{nokey,
title = {Investigating the Intelligibility of Plural Counterfactual Examples for Non-Expert Users, an Explanation User Interface Proposition and User Study},
author = {Clara Bove and Marie-Jeanne Lesot and Charles Tijus and Marcin Detyniecki},
year = {2023},
date = {2023-03-27},
booktitle = {Proceedings of the 28th International Conference on Intelligent User Interfaces, IUI'23},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {conference}
}
Krco, Natasa; Laugel, Thibault; Loubes, Jean-Michel; Detyniecki, Marcin
When Mitigating Bias is Unfair: A Comprehensive Study on the Impact of Bias Mitigation Algorithms Journal Article
In: arXiv preprint arXiv:2302.07185, 2023.
BibTeX | Tags: Explainability & Interpretability, Fairness
@article{krco2023mitigating,
title = {When Mitigating Bias is Unfair: A Comprehensive Study on the Impact of Bias Mitigation Algorithms},
author = {Natasa Krco and Thibault Laugel and Jean-Michel Loubes and Marcin Detyniecki},
year = {2023},
date = {2023-01-01},
urldate = {2023-01-01},
journal = {arXiv preprint arXiv:2302.07185},
keywords = {Explainability & Interpretability, Fairness},
pubstate = {published},
tppubtype = {article}
}
Rychener, Yves; Renard, Xavier; Seddah, Djamé; Frossard, Pascal; Detyniecki, Marcin
On the Granularity of Explanations in Model Agnostic NLP Interpretability Journal Article
In: 2022 ECML-PKDD International Workshop on eXplainable Knowledge Discovery in Data Mining (XKDD), 2022.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@article{rychener2020sentence,
title = {On the Granularity of Explanations in Model Agnostic NLP Interpretability},
author = {Yves Rychener and Xavier Renard and Djamé Seddah and Pascal Frossard and Marcin Detyniecki},
url = {https://arxiv.org/abs/2012.13189},
year = {2022},
date = {2022-09-19},
urldate = {2020-01-01},
journal = {2022 ECML-PKDD International Workshop on eXplainable Knowledge Discovery in Data Mining (XKDD)},
abstract = {Current methods for Black-Box NLP interpretability, like LIME or SHAP, are based on altering the text to interpret by removing words and modeling the Black-Box response. In this paper, we outline limitations of this approach when using complex BERT-based classifiers: The word-based sampling produces texts that are out-of-distribution for the classifier and further gives rise to a high-dimensional search space, which can't be sufficiently explored when time or computation power is limited. Both of these challenges can be addressed by using segments as elementary building blocks for NLP interpretability. As illustration, we show that the simple choice of sentences greatly improves on both of these challenges. As a consequence, the resulting explainer attains much better fidelity on a benchmark classification task.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {article}
}
Jeyasothy, Adulam; Laugel, Thibault; Lesot, Marie-Jeanne; Marsala, Christophe; Detyniecki, Marcin
Integrating Prior Knowledge in Post-hoc Explanations Proceedings Article
In: Information Processing and Management of Uncertainty in Knowledge-Based Systems: 19th International Conference, IPMU 2022, Milan, Italy, July 11--15, 2022, Proceedings, Part II, pp. 707–719, Springer 2022.
BibTeX | Tags: Explainability & Interpretability
@inproceedings{jeyasothy2022integrating,
title = {Integrating Prior Knowledge in Post-hoc Explanations},
author = {Adulam Jeyasothy and Thibault Laugel and Marie-Jeanne Lesot and Christophe Marsala and Marcin Detyniecki},
year = {2022},
date = {2022-01-01},
booktitle = {Information Processing and Management of Uncertainty in Knowledge-Based Systems: 19th International Conference, IPMU 2022, Milan, Italy, July 11--15, 2022, Proceedings, Part II},
pages = {707--719},
organization = {Springer},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {inproceedings}
}
Laugel, Thibault; Renard, Xavier; Detyniecki, Marcin
Explaining Local Discrepancies between Image Classification Models Journal Article
In: CVPR Explainable AI for Computer Vision Workshop (XAI4CV 2022), 2022.
BibTeX | Tags: Explainability & Interpretability
@article{laugel2022,
title = {Explaining Local Discrepancies between Image Classification Models},
author = {Thibault Laugel and Xavier Renard and Marcin Detyniecki},
year = {2022},
date = {2022-01-01},
journal = {CVPR Explainable AI for Computer Vision Workshop (XAI4CV 2022)},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {article}
}
Poyiadzi, Rafael; Renard, Xavier; Laugel, Thibault; Santos-Rodriguez, Raul; Detyniecki, Marcin
Understanding surrogate explanations: the interplay between complexity, fidelity and coverage Journal Article
In: arXiv preprint arXiv:2107.04309, 2021.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@article{poyiadzi2021understanding,
title = {Understanding surrogate explanations: the interplay between complexity, fidelity and coverage},
author = {Rafael Poyiadzi and Xavier Renard and Thibault Laugel and Raul Santos-Rodriguez and Marcin Detyniecki},
url = {https://arxiv.org/abs/2107.04309},
year = {2021},
date = {2021-01-01},
journal = {arXiv preprint arXiv:2107.04309},
abstract = {This paper analyses the fundamental ingredients behind surrogate explanations to provide a better understanding of their inner workings. We start our exposition by considering global surrogates, describing the trade-off between complexity of the surrogate and fidelity to the black-box being modelled. We show that transitioning from global to local - reducing coverage - allows for more favourable conditions on the Pareto frontier of fidelity-complexity of a surrogate. We discuss the interplay between complexity, fidelity and coverage, and consider how different user needs can lead to problem formulations where these are either constraints or penalties. We also present experiments that demonstrate how the local surrogate interpretability procedure can be made interactive and lead to better explanations.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {article}
}
Vermeire, Tom; Laugel, Thibault; Renard, Xavier; Martens, David; Detyniecki, Marcin
How to choose an Explainability Method? Towards a Methodical Implementation of XAI in Practice Journal Article
In: ECML PKDD International Workshop on eXplainable Knowledge Discovery in Data Mining (ECML XKDD 2021), 2021.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@article{vermeire2021choose,
title = {How to choose an Explainability Method? Towards a Methodical Implementation of XAI in Practice},
author = {Tom Vermeire and Thibault Laugel and Xavier Renard and David Martens and Marcin Detyniecki},
url = {https://arxiv.org/abs/2107.04427},
year = {2021},
date = {2021-01-01},
journal = {ECML PKDD International Workshop on eXplainable Knowledge Discovery in Data Mining (ECML XKDD 2021)},
abstract = {Explainability is becoming an important requirement for organizations that make use of automated decision-making due to regulatory initiatives and a shift in public awareness. Various and significantly different algorithmic methods to provide this explainability have been introduced in the field, but the existing literature in the machine learning community has paid little attention to the stakeholder whose needs are rather studied in the human-computer interface community. Therefore, organizations that want or need to provide this explainability are confronted with the selection of an appropriate method for their use case. In this paper, we argue there is a need for a methodology to bridge the gap between stakeholder needs and explanation methods. We present our ongoing work on creating this methodology to help data scientists in the process of providing explainability to stakeholders. In particular, our contributions include documents used to characterize XAI methods and user requirements (shown in Appendix), which our methodology builds upon.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {article}
}
Poyiadzi, Rafael; Renard, Xavier; Laugel, Thibault; Santos-Rodriguez, Raul; Detyniecki, Marcin
On the overlooked issue of defining explanation objectives for local-surrogate explainers Journal Article
In: International Conference on Machine Learning (ICML) Workshop on Theoretic Foundation, Criticism, and Application Trend of Explainable AI, 2021.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@article{poyiadzi2021overlooked,
title = {On the overlooked issue of defining explanation objectives for local-surrogate explainers},
author = {Rafael Poyiadzi and Xavier Renard and Thibault Laugel and Raul Santos-Rodriguez and Marcin Detyniecki},
url = {https://arxiv.org/abs/2106.05810},
year = {2021},
date = {2021-01-01},
journal = {International Conference on Machine Learning (ICML) Workshop on Theoretic Foundation, Criticism, and Application Trend of Explainable AI},
abstract = {Local surrogate approaches for explaining machine learning model predictions have appealing properties, such as being model-agnostic and flexible in their modelling. Several methods exist that fit this description and share this goal. However, despite their shared overall procedure, they set out different objectives, extract different information from the black-box, and consequently produce diverse explanations, that are -- in general -- incomparable. In this work we review the similarities and differences amongst multiple methods, with a particular focus on what information they extract from the model, as this has large impact on the output: the explanation. We discuss the implications of the lack of agreement, and clarity, amongst the methods' objectives on the research and practice of explainability.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {article}
}
Renard, Xavier; Laugel, Thibault; Detyniecki, Marcin
Understanding Prediction Discrepancies in Machine Learning Classifiers Journal Article
In: arXiv preprint arXiv:2104.05467, 2021.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@article{renard2021understanding,
title = {Understanding Prediction Discrepancies in Machine Learning Classifiers},
author = {Xavier Renard and Thibault Laugel and Marcin Detyniecki},
url = {https://arxiv.org/abs/2104.05467},
year = {2021},
date = {2021-01-01},
journal = {arXiv preprint arXiv:2104.05467},
abstract = {A multitude of classifiers can be trained on the same data to achieve similar performances during test time, while having learned significantly different classification patterns. This phenomenon, which we call prediction discrepancies, is often associated with the blind selection of one model instead of another with similar performances. When making a choice, the machine learning practitioner has no understanding on the differences between models, their limits, where they agree and where they don't. But his/her choice will result in concrete consequences for instances to be classified in the discrepancy zone, since the final decision will be based on the selected classification pattern. Besides the arbitrary nature of the result, a bad choice could have further negative consequences such as loss of opportunity or lack of fairness. This paper proposes to address this question by analyzing the prediction discrepancies in a pool of best-performing models trained on the same data. A model-agnostic algorithm, DIG, is proposed to capture and explain discrepancies locally, to enable the practitioner to make the best educated decision when selecting a model by anticipating its potential undesired consequences. All the code to reproduce the experiments is available.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {article}
}
Rychener, Yves; Renard, Xavier; Seddah, Djamé; Frossard, Pascal; Detyniecki, Marcin
Quackie: A nlp classification task with ground truth explanations Journal Article
In: arXiv preprint arXiv:2012.13190, 2020.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@article{rychener2020quackie,
title = {Quackie: A nlp classification task with ground truth explanations},
author = {Yves Rychener and Xavier Renard and Djamé Seddah and Pascal Frossard and Marcin Detyniecki},
url = {https://arxiv.org/abs/2012.13190},
year = {2020},
date = {2020-01-01},
journal = {arXiv preprint arXiv:2012.13190},
abstract = {NLP Interpretability aims to increase trust in model predictions. This makes evaluating interpretability approaches a pressing issue. There are multiple datasets for evaluating NLP Interpretability, but their dependence on human provided ground truths raises questions about their unbiasedness. In this work, we take a different approach and formulate a specific classification task by diverting question-answering datasets. For this custom classification task, the interpretability ground-truth arises directly from the definition of the classification problem. We use this method to propose a benchmark and lay the groundwork for future research in NLP interpretability by evaluating a wide range of current state of the art methods.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {article}
}
Laugel, Thibault
Local Post-hoc Interpretability for Black-box Classifiers PhD Thesis
Sorbonne Université, CNRS, LIP6, F-75005 Paris, France, 2020.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@phdthesis{laugel2020local,
title = {Local Post-hoc Interpretability for Black-box Classifiers},
author = {Thibault Laugel},
url = {https://hal.archives-ouvertes.fr/tel-03002496v1},
year = {2020},
date = {2020-01-01},
school = {Sorbonne Université, CNRS, LIP6, F-75005 Paris, France},
abstract = {This thesis focuses on the field of XAI (eXplainable AI), and more particularly local post-hoc interpretability paradigm, that is to say the generation of explanations for a single prediction of a trained classifier. In particular, we study a fully agnostic context, meaning that the explanation is generated without using any knowledge about the classifier (treated as a black-box) nor the data used to train it. In this thesis, we identify several issues that can arise in this context and that may be harmful for interpretability. We propose to study each of these issues and propose novel criteria and approaches to detect and characterize them. The three issues we focus on are: the risk of generating explanations that are out of distribution; the risk of generating explanations that cannot be associated to any ground-truth instance; and the risk of generating explanations that are not local enough. These risks are studied through two specific categories of interpretability approaches: counterfactual explanations, and local surrogate models.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {phdthesis}
}
Renard, Xavier; Woloszko, Nicolas; Aigrain, Jonathan; Detyniecki, Marcin
Concept tree: High-level representation of variables for more interpretable surrogate decision trees Journal Article
In: Bank of England and King's College London joint conference on Modelling with Big Data and Machine Learning: Interpretability and Model Uncertainty, 2019.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@article{renard2019concept,
title = {Concept tree: High-level representation of variables for more interpretable surrogate decision trees},
author = {Xavier Renard and Nicolas Woloszko and Jonathan Aigrain and Marcin Detyniecki},
url = {https://arxiv.org/abs/1906.01297},
year = {2019},
date = {2019-01-01},
journal = {Bank of England and King's College London joint conference on Modelling with Big Data and Machine Learning: Interpretability and Model Uncertainty},
abstract = {Interpretable surrogates of black-box predictors trained on high-dimensional tabular datasets can struggle to generate comprehensible explanations in the presence of correlated variables. We propose a model-agnostic interpretable surrogate that provides global and local explanations of black-box classifiers to address this issue. We introduce the idea of concepts as intuitive groupings of variables that are either defined by a domain expert or automatically discovered using correlation coefficients. Concepts are embedded in a surrogate decision tree to enhance its comprehensibility. First experiments on FRED-MD, a macroeconomic database with 134 variables, show improvement in human-interpretability while accuracy and fidelity of the surrogate model are preserved.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {article}
}
Renard, Xavier; Woloszko, Nicolas; Aigrain, Jonathan; Detyniecki, Marcin
Concept tree: High-level representation of variables for more interpretable surrogate decision trees Journal Article
In: International Conference on Machine Learning (ICML) Workshop on Human In the Loop Learning (HILL), 2019.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@article{renard2019conceptb,
title = {Concept tree: High-level representation of variables for more interpretable surrogate decision trees},
author = {Xavier Renard and Nicolas Woloszko and Jonathan Aigrain and Marcin Detyniecki},
url = {https://arxiv.org/abs/1906.01297},
year = {2019},
date = {2019-01-01},
journal = {International Conference on Machine Learning (ICML) Workshop on Human In the Loop Learning (HILL)},
abstract = {Interpretable surrogates of black-box predictors trained on high-dimensional tabular datasets can struggle to generate comprehensible explanations in the presence of correlated variables. We propose a model-agnostic interpretable surrogate that provides global and local explanations of black-box classifiers to address this issue. We introduce the idea of concepts as intuitive groupings of variables that are either defined by a domain expert or automatically discovered using correlation coefficients. Concepts are embedded in a surrogate decision tree to enhance its comprehensibility. First experiments on FRED-MD, a macroeconomic database with 134 variables, show improvement in human-interpretability while accuracy and fidelity of the surrogate model are preserved.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {article}
}
Laugel, Thibault; Lesot, Marie-Jeanne; Marsala, Christophe; Renard, Xavier; Detyniecki, Marcin
The Dangers of Post-hoc Interpretability: Unjustified Counterfactual Explanations Proceedings Article
In: Twenty-Eighth International Joint Conference on Artificial Intelligence $$IJCAI-19$$, pp. 2801–2807, International Joint Conferences on Artificial Intelligence Organization 2019.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@inproceedings{laugel2019dangers,
title = {The Dangers of Post-hoc Interpretability: Unjustified Counterfactual Explanations},
author = {Thibault Laugel and Marie-Jeanne Lesot and Christophe Marsala and Xavier Renard and Marcin Detyniecki},
url = {https://arxiv.org/abs/1907.09294},
year = {2019},
date = {2019-01-01},
booktitle = {Twenty-Eighth International Joint Conference on Artificial Intelligence $$IJCAI-19$$},
pages = {2801--2807},
organization = {International Joint Conferences on Artificial Intelligence Organization},
abstract = {Post-hoc interpretability approaches have been proven to be powerful tools to generate explanations for the predictions made by a trained black-box model. However, they create the risk of having explanations that are a result of some artifacts learned by the model instead of actual knowledge from the data. This paper focuses on the case of counterfactual explanations and asks whether the generated instances can be justified, i.e. continuously connected to some ground-truth data. We evaluate the risk of generating unjustified counterfactual examples by investigating the local neighborhoods of instances whose predictions are to be explained and show that this risk is quite high for several datasets. Furthermore, we show that most state of the art approaches do not differentiate justified from unjustified counterfactual examples, leading to less useful explanations.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {inproceedings}
}
Laugel, Thibault; Lesot, Marie-Jeanne; Marsala, Christophe; Renard, Xavier; Detyniecki, Marcin
Unjustified classification regions and counterfactual explanations in machine learning Proceedings Article
In: Joint European conference on machine learning and knowledge discovery in databases, pp. 37–54, Springer 2019.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@inproceedings{laugel2019unjustified,
title = {Unjustified classification regions and counterfactual explanations in machine learning},
author = {Thibault Laugel and Marie-Jeanne Lesot and Christophe Marsala and Xavier Renard and Marcin Detyniecki},
url = {https://ecmlpkdd2019.org/downloads/paper/226.pdf},
year = {2019},
date = {2019-01-01},
booktitle = {Joint European conference on machine learning and knowledge discovery in databases},
pages = {37--54},
organization = {Springer},
abstract = {Post-hoc interpretability approaches, although powerful tools to generate explanations for predictions made by a trained black-box model, have been shown to be vulnerable to issues caused by lack of robustness of the classifier. In particular, this paper focuses on the notion of explanation justification, defined as connectedness to ground-truth data, in the context of counterfactuals. In this work, we explore the extent of the risk of generating unjustified explanations. We propose an empirical study to assess the vulnerability of classifiers and show that the chosen learning algorithm heavily impacts the vulnerability of the model. Additionally, we show that state-of-the-art post-hoc counterfactual approaches can minimize the impact of this risk by generating less local explanations.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {inproceedings}
}
Laugel, Thibault; Lesot, Marie-Jeanne; Marsala, Christophe; Detyniecki, Marcin
Issues with post-hoc counterfactual explanations: a discussion Proceedings Article
In: ICML Workshop on Human in the Loop Learning (HILL 2019), 2019.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@inproceedings{laugel2019issues,
title = {Issues with post-hoc counterfactual explanations: a discussion},
author = {Thibault Laugel and Marie-Jeanne Lesot and Christophe Marsala and Marcin Detyniecki},
url = {https://arxiv.org/abs/1906.04774},
year = {2019},
date = {2019-01-01},
booktitle = {ICML Workshop on Human in the Loop Learning (HILL 2019)},
abstract = {Counterfactual post-hoc interpretability approaches have been proven to be useful tools to generate explanations for the predictions of a trained blackbox classifier. However, the assumptions they make about the data and the classifier make them unreliable in many contexts. In this paper, we discuss three desirable properties and approaches to quantify them: proximity, connectedness and stability. In addition, we illustrate that there is a risk for post-hoc counterfactual approaches to not satisfy these properties.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {inproceedings}
}
Laugel, Thibault; Renard, Xavier; Lesot, Marie-Jeanne; Marsala, Christophe; Detyniecki, Marcin
Defining Locality for Surrogates in Post-hoc Interpretablity Proceedings Article
In: Workshop on Human Interpretability for Machine Learning (WHI)-International Conference on Machine Learning (ICML), 2018.
Abstract | Links | BibTeX | Tags: Explainability & Interpretability
@inproceedings{laugel2018defining,
title = {Defining Locality for Surrogates in Post-hoc Interpretablity},
author = {Thibault Laugel and Xavier Renard and Marie-Jeanne Lesot and Christophe Marsala and Marcin Detyniecki},
url = {https://arxiv.org/abs/1806.07498},
year = {2018},
date = {2018-01-01},
booktitle = {Workshop on Human Interpretability for Machine Learning (WHI)-International Conference on Machine Learning (ICML)},
abstract = {Local surrogate models, to approximate the local decision boundary of a black-box classifier, constitute one approach to generate explanations for the rationale behind an individual prediction made by the back-box. This paper highlights the importance of defining the right locality, the neighborhood on which a local surrogate is trained, in order to approximate accurately the local black-box decision boundary. Unfortunately, as shown in this paper, this issue is not only a parameter or sampling distribution challenge and has a major impact on the relevance and quality of the approximation of the local black-box decision boundary and thus on the meaning and accuracy of the generated explanation. To overcome the identified problems, quantified with an adapted measure and procedure, we propose to generate surrogate-based explanations for individual predictions based on a sampling centered on particular place of the decision boundary, relevant for the prediction to be explained, rather than on the prediction itself as it is classically done. We evaluate the novel approach compared to state-of-the-art methods and a straightforward improvement thereof on four UCI datasets.},
keywords = {Explainability & Interpretability},
pubstate = {published},
tppubtype = {inproceedings}
}