Search Results for author:
Found 3 papers, 0 papers with code
Unreasonable Effectiveness of Last Hidden Layer Activations for Adversarial Robustness
In this study, we show both mathematically and experimentally that using some widely known activation functions in the output layer of the model with high temperature values has the effect of zeroing out the gradients for both targeted and untargeted attack cases, preventing attackers from exploiting the model's loss function to craft adversarial samples.
Exploiting epistemic uncertainty of the deep learning models to generate adversarial samples
Deep neural network architectures are considered to be robust to random perturbations.
Closeness and Uncertainty Aware Adversarial Examples Detection in Adversarial Machine Learning
While state-of-the-art Deep Neural Network (DNN) models are considered to be robust to random perturbations, it was shown that these architectures are highly vulnerable to deliberately crafted perturbations, albeit being quasi-imperceptible.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
