Evaluating the Robustness of Interpretability Methods through Explanation Invariance and Equivariance. (arXiv:2304.06715v2 [cs.LG] UPDATED)

Interpretability methods are valuable only if their explanations faithfully
describe the explained model. In this work, we consider neural networks whose
predictions are invariant under a specific symmetry group. This includes
popular architectures, ranging from convolutional to graph neural networks. Any
explanation that faithfully explains this type of model needs to be in
agreement with this invariance property. We formalize this intuition through
the notion of explanation invariance and equivariance by leveraging the
formalism from geometric deep learning. Through this …

architectures arxiv explained graph graph neural networks interpretability networks neural networks popular predictions robustness symmetry through work