BriefGPT.xyz
Oct, 2018
稀疏深度神经网络的改进对抗鲁棒性
Sparse DNNs with Improved Adversarial Robustness
HTML
PDF
Yiwen Guo, Chao Zhang, Changshui Zhang, Yurong Chen
TL;DR
本文研究了基于深度神经网络的分类模型中稀疏性与鲁棒性之间的关系，并理论和实证分析表明，适当的模型稀疏化可以提高非线性 DNN 的鲁棒性，但过度稀疏化会使模型更难抵抗对抗性样本攻击。
Abstract
deep neural networks
(DNNs) are computationally/memory-intensive and vulnerable to
adversarial attacks
, making them prohibitive in some real-world applications. By converting dense models into sparse ones, prunin
→