Matt Fredrikson

adversarial robustness (5) adversarial attack (3) neural network (2) certified robustness (2) adversarial example (2) symbolic reasoning (1) algorithmic fairness (1) adversarial learning (1) feature attribution (1) lipschitz continuity (1) ai safety (1) constrained optimization (1) model alignment (1) gaussian process (1) neural network interpretability (1) sparse optimization (1) linear regression (1) adversarial training (1) gradient descent (1) reproducing kernel hilbert space (1)