In this paper, we present a new learning framework that results in networks which are less sensitive to changes in the input, and, consequently, likely to be ...

Oct 4, 2021 · In this paper, we give a comprehensive survey on the existing studies of adversarial techniques for generating adversarial texts written by both ...

Oct 27, 2018 · Abstract:We investigate the topics of sensitivity and robustness in feedforward and convolutional neural networks.

In this paper, we develop methods to make explanations provably more robust against attacks that manipulate the input.

Abstract: Deep Neural Networks (DNNs) have gained widespread adoption, but they also exhibit post-deployment failures that pose risks to property and life.

In this paper, we give a comprehensive survey on the existing studies of adversarial techniques for generating adversarial texts written by both English and ...

Experimental results on several benchmark datasets show that, our proposed framework achieves good performance against strong adversarial at- tack methods. 1.

Based on these theoretical insights, we present three different techniques to boost robustness against manipulation: training with weight decay, smoothing ...

Experimental results indicate that the proposed method outperforms state-of-the-art sensitivity-based learning approaches with regards to robustness to ...

Machine learning models, including state-of-the-art deep neural networks, are vulnerable to small perturbations that cause unexpected classification errors.