arXiv (Cornell University) · 2018 · 89 citations · 11 references
ComputingComputer ScienceComputer SecurityMachine LearningStatistical LearningDeep LearningEnsemble MethodsGenerative ModelingNeural NetworksOptimization For Machine LearningScalable Machine LearningAdversarial RobustnessBayesian Neural NetworksMachine-learning AttacksPosterior Distribution Estimation
We present a new algorithm to train a robust neural network against adversarial attacks. Our algorithm is motivated by the following two ideas. First, although recent work has demonstrated that fusing randomness can improve the robustness of neural networks (Liu 2017), we noticed that adding noise blindly to all the layers is not the optimal way to incorporate randomness. Instead, we model randomness under the framework of Bayesian Neural Network (BNN) to formally learn the posterior distribution of models in a scalable way. Second, we formulate the mini-max problem in BNN to learn the best model distribution under adversarial attacks, leading to an adversarial-trained Bayesian neural net. Experiment results demonstrate that the proposed algorithm achieves state-of-the-art performance under strong attacks. On CIFAR-10 with VGG network, our model leads to 14\% accuracy improvement compared with adversarial training (Madry 2017) and random self-ensemble (Liu 2017) under PGD attack with $0.035$ distortion, and the gap becomes even larger on a subset of ImageNet.
11
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
Kelvin Xu, Jimmy Ba, Ryan Kiros et al. · arXiv (Cornell University) · 2015 · 7.5K citations · Full text
Intriguing properties of neural networks
Christian Szegedy, Wojciech Zaremba, Ilya Sutskever et al. · arXiv (Cornell University) · 2013 · 5.7K citations · Full text
Towards Deep Learning Models Resistant to Adversarial Attacks
Aleksander Mądry, Aleksandar Makelov, Ludwig Schmidt et al. · arXiv (Cornell University) · 2017 · 1.5K citations · Full text
Delving into Transferable Adversarial Examples and Black-box Attacks
Yanpei Liu, Xinyun Chen, Chang Liu et al. · arXiv (Cornell University) · 2016 · 551 citations · Full text