Year: 2024
Type: article
Abstract: The Rashômon Effect, applied in Explainable Machine Learning, refers to the disagreement between the explanations provided by various attribution explainers and to the dissimilarity across multiple explanations generated by a particular explainer for a single instance from the dataset (differences between feature importances and their associated signs and ranks), an undesirable outcome especially in sensitive domains such as healthcare or finance. We propose a method inspired from textual-case based reasoning for aligning explanations from various explainers in order to resolve the disagreement and dissimilarity problems. We iteratively generated a number of 100 explanations for each instance from six popular datasets, using three prevalent feature attribution explainers: LIME, Anchors and SHAP (with the variations Tree SHAP and Kernel SHAP) and consequently applied a global cluster-based aggregation strategy that quantifies alignment and reveals similarities and associations between explanations. We evaluated our method by weighting the $$\:k$$ -NN algorithm with agreed feature overlap explanation weights and compared it to a non-weighted $$\:k$$ -NN predictor, having as task binary classification. Also, we compared the results of the weighted $$\:k$$ -NN algorithm using aggregated feature overlap explanation weights to the weighted $$\:k$$ -NN algorithm using weights produced by a single explanation method (either LIME, SHAP or Anchors). Our global alignment method benefited the most from a hybridization with feature importance scores (information gain), that was essential for acquiring a more accurate estimate of disagreement, for enabling explainers to reach a consensus across multiple explanations and for supporting effective model learning through improved classification performance.
Source: Artificial Intelligence Review
Authors: Oana Mitruț, Gabriela Moise, Alin Moldoveanu, Florica Moldoveanu, Marius Leordeanu, Livia Petrescu
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $3,090
Award: G6703262589
OpenAlex ID: W4403567525