Adversarial Training of Two-Layer Polynomial and ReLU Activation Networks via Convex Optimization

AAML
Re-assign community