Search Results for author:
Found 7 papers, 2 papers with code
Prediction Risk and Estimation Risk of the Ridgeless Least Squares Estimator under General Assumptions on Regression Errors
In recent years, there has been a significant growth in research focusing on minimum $\ell_2$ norm (ridgeless) interpolation least squares estimators.
Towards Better Understanding of Training Certifiably Robust Models against Adversarial Examples
We identify another key factor that influences the performance of certifiable training: \textit{smoothness of the loss landscape}.
Implicit Jacobian regularization weighted with impurity of probability output
The eigendecomposition provides a simple relation between the eigenvalues of the low-dimensional matrix and the impurity of the probability output.
Bridged Adversarial Training
Adversarial robustness is considered as a required property of deep neural networks.
GradDiv: Adversarial Robustness of Randomized Neural Networks via Gradient Diversity Regularization
Our experiments on MNIST, CIFAR10, and STL10 show that our proposed GradDiv regularizations improve the adversarial robustness of randomized neural networks against a variety of state-of-the-art attack methods.
Loss Landscape Matters: Training Certifiably Robust Models with Favorable Loss Landscape
Certifiable training minimizes an upper bound on the worst-case loss over the allowed perturbation, and thus the tightness of the upper bound is an important factor in building certifiably robust models.
Lipschitz-Certifiable Training with a Tight Outer Bound
Our certifiable training algorithm provides a tight propagated outer bound by introducing the box constraint propagation (BCP), and it efficiently computes the worst logit over the outer bound.
