Good Counterfactuals and Where to Find Them: A Case-Based Technique for Generating Counterfactuals for Explainable AI (XAI)

Recently, a groundswell of research has identified the use of counterfactual\nexplanations as a potentially significant solution to the Explainable AI (XAI)\nproblem. It is argued that (a) technically, these counterfactual cases can be\ngenerated by permuting problem-features until a class change is found, (b)\npsychologically, they are much more causally informative than factual\nexplanations, (c) legally, they are GDPR-compliant. However, there are issues\naround the finding of good counterfactuals using current techniques (e.g.\nsparsity and plausibility). We show that many commonly-used datasets appear to\nhave few good counterfactuals for explanation purposes. So, we propose a new\ncase based approach for generating counterfactuals using novel ideas about the\ncounterfactual potential and explanatory coverage of a case-base. The new\ntechnique reuses patterns of good counterfactuals, present in a case-base, to\ngenerate analogous counterfactuals that can explain new problems and their\nsolutions. Several experiments show how this technique can improve the\ncounterfactual potential and explanatory coverage of case-bases that were\npreviously found wanting.\n