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.
Learning More Robust Features with Adversarial Training
Department of Computer Science and Technology, Chengdu University of Technology（成都理工大学计算机科学与技术学院）；Department of Geological Engineering, Chengdu University of Technology（成都理工大学地质工程系）；Department of Information and Computing Sciences, Chengdu University of Technology（成都理工大学信息与计算科学系）；Geomathematics Key Laboratory of Sichuan Province（四川省数学地质重点实验室）
Adversarial training provides a means of regularizing supervised learning algorithms while virtual adversarial training is able to extend supervised learning algorithms to the semi-supervised setting. However, both methods require making small perturbations to numerous entries of the input vector, which is inappropriate for sparse high-dimensional inputs such as one-hot word representations. We extend adversarial and virtual adversarial training to the text domain by applying perturbations to the word embeddings in a recurrent neural network rather than to the original input itself. The propos…
University of Tübingen（图宾根大学）；Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）；ETH Zürich（苏黎世联邦理工学院）
Algorithmic recourse seeks to provide actionable recommendations for individuals to overcome unfavorable classification outcomes from automated decision-making systems. Recourse recommendations should ideally be robust to reasonably small uncertainty in the features of the individual seeking recourse. In this work, we formulate the adversarially robust recourse problem and show that recourse methods that offer minimally costly recourse fail to be robust. We then present methods for generating adversarially robust recourse for linear and for differentiable classifiers. Finally, we show that reg…
Adversarial pruning compresses models while preserving robustness. Current methods require access to adversarial examples during pruning. This significantly hampers training efficiency. Moreover, as new adversarial attacks and training methods develop at a rapid rate, adversarial pruning methods need to be modified accordingly to keep up. In this work, we propose a novel framework to prune a previously trained robust neural network while maintaining adversarial robustness, without further generating adversarial examples. We leverage concurrent self-distillation and pruning to preserve knowledg…
Beihang University（北京航空航天大学）；University of Ottawa（渥太华大学）
Effective regularization techniques are highly desired in deep learning for alleviating overfitting and improving generalization. This work proposes a new regularization scheme, based on the understanding that the flat local minima of the empirical risk cause the model to generalize better. This scheme is referred to as adversarial model perturbation (AMP), where instead of directly minimizing the empirical risk, an alternative "AMP loss" is minimized via SGD. Specifically, the AMP loss is obtained from the empirical risk by applying the "worst" norm-bounded perturbation on each point in the p…
University of Southern California（南加州大学）；University of Cambridge（剑桥大学）；Carnegie Mellon University（卡内基梅隆大学）
Deep neural networks are often overparameterized and may not easily achieve model generalization. Adversarial training has shown effectiveness in improving generalization by regularizing the change of loss on top of adversarially chosen perturbations. The recently proposed sharpness-aware minimization (SAM) algorithm conducts adversarial weight perturbation, encouraging the model to converge to a flat minima. SAM finds a common adversarial weight perturbation per-batch. Although per-instance adversarial weight perturbations are stronger adversaries and can potentially lead to better generaliza…
Balance, Imbalance, and Rebalance: Understanding Robust Overfitting from a Minimax Game Perspective
Department of Engineering, University of Cambridge（剑桥大学工程系）；School of Mathematical Sciences, Peking University（北京大学数学科学学院）；Institute for Artificial Intelligence, Peking University（北京大学人工智能研究院）；Peng Cheng Laboratory（鹏城实验室）
University of Toronto（多伦多大学）；Vector Institute（向量研究所）
Do all adversarial examples have the same consequences? An autonomous driving system misclassifying a pedestrian as a car may induce a far more dangerous -- and even potentially lethal -- behavior than, for instance, a car as a bus. In order to better tackle this important problematic, we introduce the concept of hierarchical adversarial robustness. Given a dataset whose classes can be grouped into coarse-level labels, we define hierarchical adversarial examples as the ones leading to a misclassification at the coarse level. To improve the resistance of neural networks to hierarchical attacks,…
Google Research（谷歌研究院）；University of California Merced（加州大学默塞德分校）；Waymo；Yonsei University（延世大学）
Recent years have witnessed the rapid progress of generative adversarial networks (GANs). However, the success of the GAN models hinges on a large amount of training data. This work proposes a regularization approach for training robust GAN models on limited data. We theoretically show a connection between the regularized loss and an f-divergence called LeCam-divergence, which we find is more robust under limited training data. Extensive experiments on several benchmark datasets demonstrate that the proposed regularization scheme 1) improves the generalization performance and stabilizes the le…
Microsoft Research（微软研究院）；Ochanomizu University（御茶水女子大学）；University of Waterloo（滑铁卢大学）；Vector Institute（向量研究所）
We address the problem of enhancing model robustness through regularization. Specifically, we focus on methods that regularize the model posterior difference between clean and noisy inputs. Theoretically, we provide a connection of two recent methods, Jacobian Regularization and Virtual Adversarial Training, under this framework. Additionally, we generalize the posterior differential regularization to the family of $f$-divergences and characterize the overall regularization framework in terms of Jacobian matrix. Empirically, we systematically compare those regularizations and standard BERT tra…
Adversarial robustness of deep neural networks has been actively investigated. However, most existing defense approaches are limited to a specific type of adversarial perturbations. Specifically, they often fail to offer resistance to multiple attack types simultaneously, i.e., they lack multi-perturbation robustness. Furthermore, compared to image recognition problems, the adversarial robustness of video recognition models is relatively unexplored. While several studies have proposed how to generate adversarial videos, only a handful of approaches about defense strategies have been published …
Spiking neural networks (SNNs) attract great attention due to their low power consumption, low latency, and biological plausibility. As they are widely deployed in neuromorphic devices for low-power brain-inspired computing, security issues become increasingly important. However, compared to deep neural networks (DNNs), SNNs currently lack specifically designed defense methods against adversarial attacks. Inspired by neural membrane potential oscillation, we propose a novel neural model that incorporates the bio-inspired oscillation mechanism to enhance the security of SNNs. Our experiments sh…
NeuroEvolution automates the generation of Artificial Neural Networks through the application of techniques from Evolutionary Computation. The main goal of these approaches is to build models that maximize predictive performance, sometimes with an additional objective of minimizing computational complexity. Although the evolved models achieve competitive results performance-wise, their robustness to adversarial examples, which becomes a concern in security-critical scenarios, has received limited attention. In this paper, we evaluate the adversarial robustness of models found by two prominent …
In this work, we study two self-play training schemes, Chainer and Pool, and show they lead to improved agent performance in Atari Pong compared to a standard DQN agent -- trained against the built-in Atari opponent. To measure agent performance, we define a robustness metric that captures how difficult it is to learn a strategy that beats the agent's learned policy. Through playing past versions of themselves, Chainer and Pool are able to target weaknesses in their policies and improve their resistance to attack. Agents trained using these methods score well on our robustness metric and can e…
Comparing the Robustness of Modern No-Reference Image- and Video-Quality Metrics to Adversarial Attacks
MSU Institute for Artificial Intelligence（莫斯科国立大学人工智能研究所）；ISP RAS Research Center for Trusted Artificial Intelligence（俄罗斯科学院系统编程研究所可信人工智能研究中心）；Lomonosov Moscow State University（罗蒙诺索夫莫斯科国立大学）
Defense That Attacks: How Robust Models Become Better Attackers
Department of Statistics and Data Science, MBZUAI, Abu Dhabi, UAE（MBZUAI统计与数据科学系）；Cyber-Physical Systems Lab, Egypt Japan University of Science and Technology, Alexandria, Egypt（埃及日本科学技术大学智能物理系统实验室）；Faculty of Engineering, Alexandria University, Alexandria, Egypt（亚历山大大学工程学院）
University of Arizona（亚利桑那大学）；University of South Florida（南佛罗里达大学）
Deep Learning (DL)-based malware detectors are increasingly adopted for early detection of malicious behavior in cybersecurity. However, their sensitivity to adversarial malware variants has raised immense security concerns. Generating such adversarial variants by the defender is crucial to improving the resistance of DL-based malware detectors against them. This necessity has given rise to an emerging stream of machine learning research, Adversarial Malware example Generation (AMG), which aims to generate evasive adversarial malware variants that preserve the malicious functionality of a give…
Southwest University（西南大学）；Deakin University（迪肯大学）；University of Macau（澳门大学）
Deep learning models are known to be vulnerable to adversarial examples that are elaborately designed for malicious purposes and are imperceptible to the human perceptual system. Autoencoder, when trained solely over benign examples, has been widely used for (self-supervised) adversarial detection based on the assumption that adversarial examples yield larger reconstruction errors. However, because lacking adversarial examples in its training and the too strong generalization ability of autoencoder, this assumption does not always hold true in practice. To alleviate this problem, we explore ho…
