1 paper
Faezeh Labbaf, Tomáš Kolárik, Martin Blicha +3
We present a novel logic-based concept called Space Explanations for classifying neural networks that gives provable guarantees of the behavior of the network in continuous areas o…