Régression Logistique à Noyau Équivalente à un Réseau de Neurones Interprétable

This article focuses on approximating an interpretable neural network with kernel logistic regression. We introduce a new kernel that directly stems from the architecture of the neural network. The decision rule resulting from a logistic regression applied to this kernel is modeled as an additive decomposition of univariate functions and is therefore interpretable. Finally, this method achieves predictive performance similar to that of a neural network but with the advantage of offering convergence guarantees and thus result uniqueness.