paper-with-me

Papers

Improving Resistance to Adversarial Deformations by Regularizing Gradients

2020-08-29 · Pengfei Xia, Bin Li

Improving the resistance of deep neural networks against adversarial attacks is important for deploying models to realistic applications. However, most defense methods are designed to defend against intensity perturbations and ignore location perturbations, which should be equally important for deep model security. In this paper, we focus on adversarial deformations, a typical class of location perturbations, and propose a flow gradient regularization to improve the resistance of models. Theoretically, we prove that, compared with input gradient regularization, regularizing flow gradients is able to get a tighter bound. Over multiple datasets, architectures, and adversarial deformations, our empirical results indicate that models trained with flow gradients can acquire a better resistance than trained with input gradients with a large margin, and also better than adversarial training. Moreover, compared with directly training with adversarial deformations, our method can achieve better results in unseen attacks, and combining these two methods can improve the resistance further.

📄 PDF Abstract BibTeX arXiv:2008.12997

Code (1)

xpf/Flow-Gradient-Regularization 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Improving Back-Propagation by Adding an Adversarial Gradient

2015-10-14 · Arild Nøkland

The back-propagation algorithm is widely used for learning in artificial neural networks. A challenge in machine learning is to create models that generalize to new data samples not seen in the training data. Recently, a…

BIG-bench Machine LearningGeneral Classification

Characterizing Model Robustness via Natural Input Gradients

2024-09-30 · Adrián Rodríguez-Muñoz, Tongzhou Wang, Antonio Torralba

Adversarially robust models are locally smooth around each data sample so that small perturbations cannot drastically change model outputs. In modern systems, such smoothness is usually obtained via Adversarial Training,…

model

Improving the Adversarial Robustness and Interpretability of Deep Neural Networks by Regularizing their Input Gradients

2017-11-26 · Andrew Slavin Ross, Finale Doshi-Velez

Deep neural networks have proven remarkably effective at solving many classification problems, but have been criticized recently for two major weaknesses: the reasons behind their predictions are uninterpretable, and the…

Adversarial Robustness

Towards Robust Training of Neural Networks by Regularizing Adversarial Gradients

2018-05-23 · Fuxun Yu, Zirui Xu, Yanzhi Wang, ChenChen Liu 외

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 pos…

Neural Networks with Structural Resistance to Adversarial Attacks

2018-09-25 · ICLR 2019 5 · Luca de Alfaro

In adversarial attacks to machine-learning classifiers, small perturbations are added to input that is correctly classified. The perturbations yield adversarial examples, which are virtually indistinguishable from the un…