ANNs and Unifying Explanations: Reply to Erasmus, Brunet, and Fisher

In a recent article, Erasmus, Brunet, and Fisher (2021) argue that Artificial Neural Networks (ANNs) are explainable. They survey four influential accounts of explanation: the Deductive-Nomological model, Inductive-Statistical Causal-Mechanical New-Mechanist model. that, on each these accounts, features make something an explanation is invariant with regard to complexity explanans explanandum. Therefore, they conclude, ANNs (and other Machine Learning models) does not them less this reply, it argued Erasmus et al. left out one account from their discussion: unificationist It sensitive complexity. explainable.It emphasized al.’s general strategy correct. The literature explainable Intelligence can benefit by drawing philosophical explanation. However, do settle problem whether because unanimously declare