首页/论文/2009.08058

arXiv 2009.080582020-09cs.LGcs.CVstat.ML

MultAV: Multiplicative Adversarial Videos

Johns Hopkins University（约翰斯·霍普金斯大学）

Shao-Yuan Lo, Vishal M. Patel

备注：: Accepted at IEEE International Conference on Advanced Video and Signal-based Surveillance (AVSS) 2021

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

英文摘要相关论文

英文摘要：

The majority of adversarial machine learning research focuses on additive attacks, which add adversarial perturbation to input data. On the other hand, unlike image recognition problems, only a handful of attack approaches have been explored in the video domain. In this paper, we propose a novel attack method against video recognition models, Multiplicative Adversarial Videos (MultAV), which imposes perturbation on video data by multiplication. MultAV has different noise distributions to the additive counterparts and thus challenges the defense methods tailored to resisting additive adversarial attacks. Moreover, it can be generalized to not only Lp-norm attacks with a new adversary constraint called ratio bound, but also different types of physically realizable attacks. Experimental results show that the model adversarially trained against additive attack is less robust to MultAV.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

继续探索

相关论文

Learning with Multiplicative Perturbations

Georgia State University（佐治亚州立大学）

Adversarial Training (AT) and Virtual Adversarial Training (VAT) are the regularization techniques that train Deep Neural Networks (DNNs) with adversarial examples generated by adding small but worst-case perturbations to input examples. In this paper, we propose xAT and xVAT, new adversarial training algorithms, that generate \textbf{multiplicative} perturbations to input examples for robust training of DNNs. Such perturbations are much more perceptible and interpretable than their \textbf{additive} counterparts exploited by AT and VAT. Furthermore, the multiplicative perturbations can be gen…

2020-06-22cs.LGstat.ML

Toward Optimal Adversarial Policies in the Multiplicative Learning System with a Malicious Expert

University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Ecole polytechnique fédérale de Lausanne (EPFL)（洛桑联邦理工学院）；Princeton University（普林斯顿大学）

We consider a learning system based on the conventional multiplicative weight (MW) rule that combines experts' advice to predict a sequence of true outcomes. It is assumed that one of the experts is malicious and aims to impose the maximum loss on the system. The loss of the system is naturally defined to be the aggregate absolute difference between the sequence of predicted outcomes and the true outcomes. We consider this problem under both offline and online settings. In the offline setting where the malicious expert must choose its entire sequence of decisions a priori, we show somewhat sur…

2020-09-18cs.CRcs.LGcs.MA

可抵御对抗攻击的随机多臂老虎机

Stochastic Bandits Robust to Adversarial Attacks

CICS, UMass Amherst（马萨诸塞大学阿默斯特分校CICS）；CS, CityU（香港城市大学计算机科学系）；CSE, CUHK（香港中文大学计算机科学与工程系）

本文针对攻击者可篡改奖励观测的随机多臂老虎机场景，设计了攻击预算已知/未知两类场景下的鲁棒算法，给出对应遗憾上界与下界，证明算法最优性并明确其与腐败模型的差异。

2024-08-16cs.LG

Overcomplete Representations Against Adversarial Videos

Johns Hopkins University（约翰斯·霍普金斯大学）

Adversarial robustness of deep neural networks is an extensively studied problem in the literature and various methods have been proposed to defend against adversarial images. However, only a handful of defense methods have been developed for defending against attacked videos. In this paper, we propose a novel Over-and-Under complete restoration network for Defending against adversarial videos (OUDefend). Most restoration networks adopt an encoder-decoder architecture that first shrinks spatial dimension then expands it back. This approach learns undercomplete representations, which have large…

2021-06-14cs.CVcs.LGeess.IV

Neural SDE: Stabilizing Neural ODE Networks with Stochastic Noise

UCLA（加州大学洛杉矶分校）；UC Davis（加州大学戴维斯分校）；Google Research（谷歌研究院）

Neural Ordinary Differential Equation (Neural ODE) has been proposed as a continuous approximation to the ResNet architecture. Some commonly used regularization mechanisms in discrete neural networks (e.g. dropout, Gaussian noise) are missing in current Neural ODE networks. In this paper, we propose a new continuous neural network framework called Neural Stochastic Differential Equation (Neural SDE) network, which naturally incorporates various commonly used regularization mechanisms based on random noise injection. Our framework can model various types of noise injection frequently used in di…

2019-06-05cs.AIcs.CVcs.LG

Robust Attacks against Multiple Classifiers

University of California, Berkeley（加州大学伯克利分校）；Harvard University（哈佛大学）

We address the challenge of designing optimal adversarial noise algorithms for settings where a learner has access to multiple classifiers. We demonstrate how this problem can be framed as finding strategies at equilibrium in a two-player, zero-sum game between a learner and an adversary. In doing so, we illustrate the need for randomization in adversarial attacks. In order to compute Nash equilibrium, our main technical focus is on the design of best response oracles that can then be implemented within a Multiplicative Weights Update framework to boost deterministic perturbations against a se…

2019-06-06cs.CRcs.GTcs.LG

更多相关论文收起更多论文

Adversarial robustness via robust low rank representations

Google Research（谷歌研究院）；Department of Computer Science, Rutgers University（罗格斯大学计算机科学系）；Department of Computer Science, Northwestern University（西北大学计算机科学系）

Adversarial robustness measures the susceptibility of a classifier to imperceptible perturbations made to the inputs at test time. In this work we highlight the benefits of natural low rank representations that often exist for real data such as images, for training neural networks with certified robustness guarantees. Our first contribution is for certified robustness to perturbations measured in $\ell_2$ norm. We exploit low rank data representations to provide improved guarantees over state-of-the-art randomized smoothing-based approaches on standard benchmark datasets such as CIFAR-10 and C…

2020-08-01cs.DScs.LGstat.ML

一种通过扩散方程稳定的可调谐去斑神经网络

A Tunable Despeckling Neural Network Stabilized via Diffusion Equation

School of Mathematics, Harbin Institute of Technology（哈尔滨工业大学数学学院）；Helmholtz Imaging, Deutsches Elektronen-Synchrotron DESY（德国电子同步加速器亥姆霍兹成像研究所）；Fachbereich Mathematik, Universität Hamburg（汉堡大学数学系）

针对SAR图像去斑中神经网络易受扰动影响的问题，提出一种结合浅层去噪与热方程扩散正则化的可调谐网络，理论保证稳定收敛，实验在模拟、对抗及真实SAR图像上超越现有方法。

2026-09-22cs.CVeess.IV

利用乘性权重扰动提升对损坏的鲁棒性

Improving robustness to corruptions with multiplicative weight perturbations

Aalto University（阿尔托大学）；University of Helsinki（赫尔辛基大学）；University of Manchester（曼彻斯特大学）

提出乘性权重扰动数据增强方法DAMP，在不损失干净图像准确率的前提下提升DNNs对多种损坏的鲁棒性，并在多个数据集和架构上验证其有效性。

2024-12-24cs.CVcs.LG

Divide-and-Conquer Adversarial Learning for High-Resolution Image and Video Enhancement

ETH Zurich（苏黎世联邦理工学院）；KU Leuven（鲁汶大学）

This paper introduces a divide-and-conquer inspired adversarial learning (DACAL) approach for photo enhancement. The key idea is to decompose the photo enhancement process into hierarchically multiple sub-problems, which can be better conquered from bottom to up. On the top level, we propose a perception-based division to learn additive and multiplicative components, required to translate a low-quality image or video into its high-quality counterpart. On the intermediate level, we use a frequency-based division with generative adversarial network (GAN) to weakly supervise the photo enhancement…

2019-10-23cs.CVcs.LGeess.IV

Adequacy of the Gradient-Descent Method for Classifier Evasion Attacks

University of Melbourne（墨尔本大学）

Despite the wide use of machine learning in adversarial settings including computer security, recent studies have demonstrated vulnerabilities to evasion attacks---carefully crafted adversarial samples that closely resemble legitimate instances, but cause misclassification. In this paper, we examine the adequacy of the leading approach to generating adversarial samples---the gradient descent approach. In particular (1) we perform extensive experiments on three datasets, MNIST, USPS and Spambase, in order to analyse the effectiveness of the gradient-descent method against non-linear support vec…

2017-05-25cs.CRcs.LGstat.ML

高效在线量子生成对抗学习算法及其应用

Efficient Online Quantum Generative Adversarial Learning Algorithms with Applications

The University of Sydney（悉尼大学）；University of Technology Sydney（悉尼科技大学）

针对量子生成对抗学习缺乏明确算法的问题，提出融合在线学习的QMMW算法，并通过参数化量子电路与纯态纠缠测试验证其高效性。

2019-04-21cs.LGquant-ph

Robust Linear Regression: Gradient-descent, Early-stopping, and Beyond

Facebook AI Research（脸书人工智能研究院）

In this work we study the robustness to adversarial attacks, of early-stopping strategies on gradient-descent (GD) methods for linear regression. More precisely, we show that early-stopped GD is optimally robust (up to an absolute constant) against Euclidean-norm adversarial attacks. However, we show that this strategy can be arbitrarily sub-optimal in the case of general Mahalanobis attacks. This observation is compatible with recent findings in the case of classification~\cite{Vardi2022GradientMP} that show that GD provably converges to non-robust models. To alleviate this issue, we propose …

2023-01-31cs.LGstat.ML

Robust conditional GANs under missing or uncertain labels

University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；University of Washington, Seattle（西雅图华盛顿大学）；Amazon（亚马逊）

Matching the performance of conditional Generative Adversarial Networks with little supervision is an important task, especially in venturing into new domains. We design a new training algorithm, which is robust to missing or ambiguous labels. The main idea is to intentionally corrupt the labels of generated examples to match the statistics of the real data, and have a discriminator process the real and generated examples with corrupted labels. We showcase the robustness of this proposed approach both theoretically and empirically. We show that minimizing the proposed loss is equivalent to min…

2019-06-09cs.LGstat.ML

On Linear Stability of SGD and Input-Smoothness of Neural Networks

Stanford University（斯坦福大学）

The multiplicative structure of parameters and input data in the first layer of neural networks is explored to build connection between the landscape of the loss function with respect to parameters and the landscape of the model function with respect to input data. By this connection, it is shown that flat minima regularize the gradient of the model function, which explains the good generalization performance of flat minima. Then, we go beyond the flatness and consider high-order moments of the gradient noise, and show that Stochastic Gradient Descent (SGD) tends to impose constraints on these…

2021-11-28cs.LG

理解并利用双人博弈方法减少数据增强的类依赖效应

Understanding and Reducing the Class-Dependent Effects of Data Augmentation with A Two-Player Game Approach

Shanghai Jiao Tong University（上海交通大学）；Duke Kunshan University（杜克大学）

本文针对数据增强在多类分类中产生的不公平类依赖效应，提出类依赖乘性权重方法CLAM，将分类器训练表述为对抗性双人博弈并设计收敛的乘性权重算法，实验证明该方法能在几乎不影响平均准确率的情况下使各类别性能分布更公平。

2025-06-30cs.AIcs.CVcs.CY