@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}

}

@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}

}

@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}

}

@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}

}

@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}

}

@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}

}

@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}

}

@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}

}

@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}

}

@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}

}

@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}

}