Random Projections for Improved Adversarial Robustness
This addresses the issue of adversarial vulnerabilities in neural networks for security-critical applications, presenting an incremental improvement over existing methods.
The paper tackles the problem of improving neural network robustness to adversarial attacks by proposing two training techniques that use random projections of inputs, achieving improved robustness without specifying concrete numerical results.
We propose two training techniques for improving the robustness of Neural Networks to adversarial attacks, i.e. manipulations of the inputs that are maliciously crafted to fool networks into incorrect predictions. Both methods are independent of the chosen attack and leverage random projections of the original inputs, with the purpose of exploiting both dimensionality reduction and some characteristic geometrical properties of adversarial perturbations. The first technique is called RP-Ensemble and consists of an ensemble of networks trained on multiple projected versions of the original inputs. The second one, named RP-Regularizer, adds instead a regularization term to the training objective.