no code implementations • 30 Apr 2024 • Olivier Letoffe, Xuanxiang Huang, Joao Marques-Silva
Recent work uncovered examples of classifiers for which SHAP scores yield misleading feature attributions.
no code implementations • 30 Sep 2023 • Xuanxiang Huang, Joao Marques-Silva
Recent work demonstrated the inadequacy of Shapley values for explainable artificial intelligence (XAI).
Explainable artificial intelligence Explainable Artificial Intelligence (XAI)
no code implementations • 6 Sep 2023 • Xuanxiang Huang, Joao Marques-Silva
This earlier work devised a brute-force approach to identify Boolean functions, defined on small numbers of features, and also associated instances, which displayed such inadequacy-revealing issues, and so served as evidence to the inadequacy of Shapley values for rule-based explainability.
no code implementations • 27 Jun 2023 • Joao Marques-Silva, Xuanxiang Huang
Explainable artificial intelligence (XAI) aims to help human decision-makers in understanding complex machine learning (ML) models.
Explainable artificial intelligence Explainable Artificial Intelligence (XAI) +1
no code implementations • 5 Jun 2023 • Xuanxiang Huang, Joao Marques-Silva
In contrast with ad-hoc methods for eXplainable Artificial Intelligence (XAI), formal explainability offers important guarantees of rigor.
Explainable artificial intelligence Explainable Artificial Intelligence (XAI)
no code implementations • 16 Feb 2023 • Xuanxiang Huang, Joao Marques-Silva
This paper develops a rigorous argument for why the use of Shapley values in explainable AI (XAI) will necessarily yield provably misleading information about the relative importance of features for predictions.
no code implementations • 12 Dec 2022 • Yacine Izza, Xuanxiang Huang, Alexey Ignatiev, Nina Narodytska, Martin C. Cooper, Joao Marques-Silva
One solution is to consider intrinsic interpretability, which does not exhibit the drawback of unsoundness.
1 code implementation • 27 Oct 2022 • Xuanxiang Huang, Martin C. Cooper, Antonio Morgado, Jordi Planes, Joao Marques-Silva
Given a machine learning (ML) model and a prediction, explanations can be defined as sets of features which are sufficient for the prediction.
no code implementations • 15 Feb 2022 • Xuanxiang Huang, Joao Marques-Silva
In contrast, this paper shows that for a number of families of classifiers, FMP is in NP.
no code implementations • 4 Jul 2021 • Xuanxiang Huang, Yacine Izza, Alexey Ignatiev, Martin C. Cooper, Nicholas Asher, Joao Marques-Silva
Knowledge compilation (KC) languages find a growing number of practical uses, including in Constraint Programming (CP) and in Machine Learning (ML).
1 code implementation • 2 Jun 2021 • Xuanxiang Huang, Yacine Izza, Alexey Ignatiev, Joao Marques-Silva
Recent work has shown that not only decision trees (DTs) may not be interpretable but also proposed a polynomial-time algorithm for computing one PI-explanation of a DT.