Ensemble of Counterfactual Explainers

Riccardo Guidotti, Salvatore Ruggieri
8/29/2023
cs.AIcs.LG

No official code found. Best implementation: riccotti/ECE (3 stars, last commit 2022-05-27).

Abstract

In eXplainable Artificial Intelligence (XAI), several counterfactual explainers have been proposed, each focusing on some desirable properties of counterfactual instances: minimality, actionability, stability, diversity, plausibility, discriminative power. We propose an ensemble of counterfactual explainers that boosts weak explainers, which provide only a subset of such properties, to a powerful method covering all of them. The ensemble runs weak explainers on a sample of instances and of features, and it combines their results by exploiting a diversity-driven selection function. The method is model-agnostic and, through a wrapping approach based on autoencoders, it is also data-agnostic.

View on arXivView PDF

Code Implementations(1)

riccotti/ECECommunity70%

Ensemble of Counterfactual Explainers

32Shell, PythonMay 31, 20214 years ago

Similar Papers

Achieving Diversity in Counterfactual Explanations: a Review and Discussion

89% match2023

Beyond One-Size-Fits-All: Adapting Counterfactual Explanations to User Objectives

86% match2024

Introducing User Feedback-based Counterfactual Explanations (UFCE)

85% match1 repo2024

Overlap Number of Balls Model-Agnostic CounterFactuals (ONB-MACF): A Data-Morphology-based Counterfactual Generation Method for Trustworthy Artificial Intelligence

85% match2024

Exploring Energy Landscapes for Minimal Counterfactual Explanations: Applications in Cybersecurity and Beyond

85% match2025

On the Hardness of Computing Counterfactual and Semifactual Explanations in XAI

85% match2026

Cite this paper

@article{guidotti2023ensemble,
  title  = {Ensemble of Counterfactual Explainers},
  author = {Riccardo Guidotti and Salvatore Ruggieri},
  year   = {2023},
  eprint = {2308.15194},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2308.15194v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2308.15194.svg)](https://codeofpaper.com/paper/2308.15194)

Discussion