Abstract We propose answer-set programs that specify and compute counterfactual interventions on entities are input a classification model. In relation to the outcome of model, resulting serve as basis for definition computation causality-based explanation scores feature values in entity under classification, namely responsibility . The approach can be applied with black-box models, also models...

Previous work has shown that both declarative and non-declarative strategies can be engaged in probabilistic classification learning. With respect to the neural correlates of these strategies, earlier studies have focused on the classification process itself. In the present experiment, we asked whether the feedback for classification performance is processed differently by declarative and non-d...