Towards Robust Training of Neural Networks by Regularizing Adversarial Gradients
This addresses the problem of adversarial vulnerability in neural networks for AI security applications, offering a novel approach without using adversarial examples, but it appears incremental as it builds on existing gradient-based defenses.
The paper tackles the susceptibility of neural networks to adversarial examples by proposing a robust training method that regulates adversarial gradients, achieving optimal accuracy against FGSM and C&W attacks on MNIST, Cifar10, and Google Speech Command datasets.
In recent years, neural networks have demonstrated outstanding effectiveness in a large amount of applications.However, recent works have shown that neural networks are susceptible to adversarial examples, indicating possible flaws intrinsic to the network structures. To address this problem and improve the robustness of neural networks, we investigate the fundamental mechanisms behind adversarial examples and propose a novel robust training method via regulating adversarial gradients. The regulation effectively squeezes the adversarial gradients of neural networks and significantly increases the difficulty of adversarial example generation.Without any adversarial example involved, the robust training method could generate naturally robust networks, which are near-immune to various types of adversarial examples. Experiments show the naturally robust networks can achieve optimal accuracy against Fast Gradient Sign Method (FGSM) and C\&W attacks on MNIST, Cifar10, and Google Speech Command dataset. Moreover, our proposed method also provides neural networks with consistent robustness against transferable attacks.