This is your work, valued
expressive-losses. Expressive Losses for Verified Robustness via Convex Combinations
ibpr. IBP Regularization for Verified Adversarial Robustness via Branch-and-Bound
favex. Faster Verified Explanations for Neural Networks
