Adversarially Robust Learning via Entropic Regularization.
Adversarially Robust Learning via Entropic Regularization.
复制标题
DOI：
10.3389/frai.2021.780843
复制
发表时间：
2021
影响因子：
4
通讯作者：
Hegde C
中科院分区：
文献类型：
--
作者：
Jagatap G;Joshi A;Chowdhury AB;Garg S;Hegde C
在本文中，我们提出了一个新的算法家族，ATENT，用于训练对抗鲁棒的深度神经网络。我们制定了一个新的损失函数，配备了额外的熵正则化。我们的损失函数考虑了对抗样本的贡献，这些样本是从数据空间中专门设计的分布中提取的，该分布将高概率分配给具有高损失的点和训练样本的近邻。我们提出的算法优化了这种损失，以寻求损失景观的对抗性强的山谷。与MNIST和CIFAR-10等基准数据集上的几种最先进的鲁棒学习方法相比，我们的方法在鲁棒分类精度方面具有竞争力（或更好）的性能。
In this paper we propose a new family of algorithms, ATENT, for training adversarially robust deep neural networks. We formulate a new loss function that is equipped with an additional entropic regularization. Our loss function considers the contribution of adversarial samples that are drawn from a specially designed distribution in the data space that assigns high probability to points with high loss and in the immediate neighborhood of training samples. Our proposed algorithms optimize this loss to seek adversarially robust valleys of the loss landscape. Our approach achieves competitive (or better) performance in terms of robust classification accuracy as compared to several state-of-the-art robust learning approaches on benchmark datasets such as MNIST and CIFAR-10.
DOI：
10.1088/1742-5468/ab39d9
发表时间：
2019-12-01
影响因子：
2.4
作者：
Chaudhari, Pratik;Choromanska, Anna;Zecchina, Riccardo
通讯作者：
Zecchina, Riccardo
影响因子：
8
作者：
Biggio, Battista;Roli, Fabio
通讯作者：
Roli, Fabio
影响因子：
22.7
作者：
Krizhevsky, Alex;Sutskever, Ilya;Hinton, Geoffrey E.
通讯作者：
Hinton, Geoffrey E.