Graph Edits for Counterfactual Explanations: A comparative study

arXiv:2401.11609v2 Announce Type: replace
Abstract: Counterfactuals have been established as a popular explainability technique which leverages a set of minimal edits to alter the prediction of a classifier. When considering conceptual counterfactuals on images, the edits requested should correspond to salient concepts present in the input data. At the same time, conceptual distances are defined by knowledge graphs, ensuring the optimality of conceptual edits. In this work, we extend previous endeavors on graph edits as counterfactual explanations by conducting a …

abstract arxiv classifier concepts counterfactual cs.ai cs.lg data explainability graph images popular prediction set study type