Evaluating the explanations given by post-hoc XAI approaches on tabular data is a challenging prospect, since the subjective judgement of explanations of tabular relations is non trivial in contrast to e.g. the judgement of image heatmap explanations. In order to quantify XAI performance on categorical tabular data, where feature relationships can often be described by Boolean functions, we propose an evaluation setting through generation of synthetic datasets. To create gold standard explanations, we present a definition of feature relevance in Boolean functions. In the proposed setting we evaluate eight state-of-the-art XAI approaches and gain novel insights into XAI performance on categorical tabular data. We find that the investigated approaches often fail to faithfully explain even basic relationships within categorical data.
%0 Journal Article
%1 tritscher2020evaluation
%A Tritscher, Julian
%A Ring, Markus
%A Schlör, Daniel
%A Hettinger, Lena
%A Hotho, Andreas
%B International Symposium on Methodologies for Intelligent Systems
%D 2020
%K author:schloer dataset evaluation from:tritsch myown research_xai xai
%T Evaluation of post-hoc XAI approaches through synthetic tabular data
%X Evaluating the explanations given by post-hoc XAI approaches on tabular data is a challenging prospect, since the subjective judgement of explanations of tabular relations is non trivial in contrast to e.g. the judgement of image heatmap explanations. In order to quantify XAI performance on categorical tabular data, where feature relationships can often be described by Boolean functions, we propose an evaluation setting through generation of synthetic datasets. To create gold standard explanations, we present a definition of feature relevance in Boolean functions. In the proposed setting we evaluate eight state-of-the-art XAI approaches and gain novel insights into XAI performance on categorical tabular data. We find that the investigated approaches often fail to faithfully explain even basic relationships within categorical data.
@article{tritscher2020evaluation,
abstract = {Evaluating the explanations given by post-hoc XAI approaches on tabular data is a challenging prospect, since the subjective judgement of explanations of tabular relations is non trivial in contrast to e.g. the judgement of image heatmap explanations. In order to quantify XAI performance on categorical tabular data, where feature relationships can often be described by Boolean functions, we propose an evaluation setting through generation of synthetic datasets. To create gold standard explanations, we present a definition of feature relevance in Boolean functions. In the proposed setting we evaluate eight state-of-the-art XAI approaches and gain novel insights into XAI performance on categorical tabular data. We find that the investigated approaches often fail to faithfully explain even basic relationships within categorical data.},
added-at = {2020-03-13T03:10:49.000+0100},
author = {Tritscher, Julian and Ring, Markus and Schlör, Daniel and Hettinger, Lena and Hotho, Andreas},
biburl = {https://www.bibsonomy.org/bibtex/25ab26432231e48433998f369f2af9d15/dmir},
booktitle = {International Symposium on Methodologies for Intelligent Systems},
interhash = {1e0ea031dac2d90618e587e3f9d958c8},
intrahash = {5ab26432231e48433998f369f2af9d15},
keywords = {author:schloer dataset evaluation from:tritsch myown research_xai xai},
note = {Accepted but not published},
organization = {Springer},
timestamp = {2024-01-18T10:31:52.000+0100},
title = {Evaluation of post-hoc XAI approaches through synthetic tabular data},
year = 2020
}