Viewing the process of generating counterfactuals as a source of knowledge: a new approach for explaining classifiers
This is an incremental approach for improving explainable AI methods, potentially benefiting users who need interpretable machine learning models.
The paper tackles the problem of explaining classifier decisions by proposing to treat the process of generating counterfactuals as a source of knowledge that can be stored and reused, illustrated with additive models and naive Bayes classifiers.
There are now many explainable AI methods for understanding the decisions of a machine learning model. Among these are those based on counterfactual reasoning, which involve simulating features changes and observing the impact on the prediction. This article proposes to view this simulation process as a source of creating a certain amount of knowledge that can be stored to be used, later, in different ways. This process is illustrated in the additive model and, more specifically, in the case of the naive Bayes classifier, whose interesting properties for this purpose are shown.