Training Provably Robust Models by Polyhedral Envelope Regularization.
Sign Up to like & getrecommendations! Published in 2021 at "IEEE transactions on neural networks and learning systems"
DOI: 10.1109/tnnls.2021.3111892
Abstract: Training certifiable neural networks enables us to obtain models with robustness guarantees against adversarial attacks. In this work, we introduce a framework to obtain a provable adversarial-free region in the neighborhood of the input data… read more here.
Keywords: polyhedral envelope; robust models; training provably; envelope regularization ... See more keywords