中文
相关论文

相关论文: Boosting Ray Search Procedure of Hard-label Attack…

200 篇论文

In hard-label black-box adversarial attacks, where only the top-1 predicted label is accessible, the prohibitive query complexity poses a major obstacle to practical deployment. In this paper, we focus on optimizing a representative class…

机器学习 · 计算机科学 2025-12-25 Xinjie Xu , Shuyu Cheng , Dongwei Xu , Qi Xuan , Chen Ma

Adversarial attacks have been extensively studied in recent years since they can identify the vulnerability of deep learning models before deployed. In this paper, we consider the black-box adversarial setting, where the adversary needs to…

机器学习 · 计算机科学 2022-03-15 Yinpeng Dong , Shuyu Cheng , Tianyu Pang , Hang Su , Jun Zhu

We consider the black-box adversarial setting, where the adversary has to generate adversarial perturbations without access to the target models to compute gradients. Previous methods tried to approximate the gradient either by using a…

机器学习 · 计算机科学 2020-07-28 Shuyu Cheng , Yinpeng Dong , Tianyu Pang , Hang Su , Jun Zhu

Deep neural networks are vulnerable to adversarial attacks. Among different attack settings, the most challenging yet the most practical one is the hard-label setting where the attacker only has access to the hard-label output (prediction…

机器学习 · 计算机科学 2020-09-08 Jinghui Chen , Quanquan Gu

We consider the hard label based black box adversarial attack setting which solely observes predicted classes from the target model. Most of the attack methods in this setting suffer from impractical number of queries required to achieve a…

机器学习 · 计算机科学 2024-03-12 Jeonghwan Park , Paul Miller , Niall McLaughlin

We study the most practical problem setup for evaluating adversarial robustness of a machine learning system with limited access: the hard-label black-box attack setting for generating adversarial examples, where limited model queries are…

机器学习 · 计算机科学 2020-02-17 Minhao Cheng , Simranjit Singh , Patrick Chen , Pin-Yu Chen , Sijia Liu , Cho-Jui Hsieh

We study the problem of generating adversarial examples in a black-box setting in which only loss-oracle access to a model is available. We introduce a framework that conceptually unifies much of the existing work on black-box attacks, and…

机器学习 · 统计学 2019-03-29 Andrew Ilyas , Logan Engstrom , Aleksander Madry

Hard-label black-box attacks, relying solely on top-1 predictions, represent one of the most challenging yet practically threat models. Despite recent progress, existing approaches face two key limitations: (1) they overlook the critical…

机器学习 · 计算机科学 2026-05-25 Jun Liu , Leo Yu Zhang , Fengpeng Li , Isao Echizen , Jiantao Zhou

A recent line of work on black-box adversarial attacks has revived the use of transfer from surrogate models by integrating it into query-based search. However, we find that existing approaches of this type underperform their potential, and…

机器学习 · 计算机科学 2022-03-17 Nicholas A. Lord , Romain Mueller , Luca Bertinetto

Deep neural networks are vulnerable to adversarial examples, which are crafted by adding human-imperceptible perturbations to original images. Most existing adversarial attack methods achieve nearly 100% attack success rates under the…

计算机视觉与模式识别 · 计算机科学 2022-04-15 Guoqiu Wang , Huanqian Yan , Ying Guo , Xingxing Wei

We present a new method for black-box adversarial attack. Unlike previous methods that combined transfer-based and scored-based methods by using the gradient or initialization of a surrogate white-box model, this new method tries to learn a…

机器学习 · 计算机科学 2020-01-07 Zhichao Huang , Tong Zhang

Decision-based methods have shown to be effective in black-box adversarial attacks, as they can obtain satisfactory performance and only require to access the final model prediction. Gradient estimation is a critical step in black-box…

机器学习 · 计算机科学 2023-10-31 Han Liu , Xingshuo Huang , Xiaotong Zhang , Qimai Li , Fenglong Ma , Wei Wang , Hongyang Chen , Hong Yu , Xianchao Zhang

We study the problem of attacking a machine learning model in the hard-label black-box setting, where no model information is revealed except that the attacker can make queries to probe the corresponding hard-label decisions. This is a very…

机器学习 · 计算机科学 2018-07-13 Minhao Cheng , Thong Le , Pin-Yu Chen , Jinfeng Yi , Huan Zhang , Cho-Jui Hsieh

This paper studies the challenging black-box adversarial attack that aims to generate adversarial examples against a black-box model by only using output feedback of the model to input queries. Some previous methods improve the query…

机器学习 · 计算机科学 2024-05-30 Shuyu Cheng , Yibo Miao , Yinpeng Dong , Xiao Yang , Xiao-Shan Gao , Jun Zhu

We focus on the problem of black-box adversarial attacks, where the aim is to generate adversarial examples for deep learning models solely based on information limited to output label~(hard label) to a queried data input. We propose a…

机器学习 · 计算机科学 2021-06-14 Satya Narayan Shukla , Anit Kumar Sahu , Devin Willmott , J. Zico Kolter

Black-box query attacks, which rely only on the output of the victim model, have proven to be effective in attacking deep learning models. However, existing black-box query attacks show low performance in a novel scenario where only a few…

计算机视觉与模式识别 · 计算机科学 2023-05-04 Xiangyuan Yang , Jie Lin , Hanlin Zhang , Xinyu Yang , Peng Zhao

This paper addresses the challenging black-box adversarial attack problem, where only classification confidence of a victim model is available. Inspired by consistency of visual saliency between different vision models, a surrogate model is…

密码学与安全 · 计算机科学 2020-10-23 Jiancheng Yang , Yangzhou Jiang , Xiaoyang Huang , Bingbing Ni , Chenglong Zhao

We propose the first general-purpose gradient-based attack against transformer models. Instead of searching for a single adversarial example, we search for a distribution of adversarial examples parameterized by a continuous-valued matrix,…

计算与语言 · 计算机科学 2021-04-29 Chuan Guo , Alexandre Sablayrolles , Hervé Jégou , Douwe Kiela

Solving for adversarial examples with projected gradient descent has been demonstrated to be highly effective in fooling the neural network based classifiers. However, in the black-box setting, the attacker is limited only to the query…

机器学习 · 计算机科学 2022-10-19 Seungyong Moon , Gaon An , Hyun Oh Song

Unlike the white-box counterparts that are widely studied and readily accessible, adversarial examples in black-box settings are generally more Herculean on account of the difficulty of estimating gradients. Many methods achieve the task by…

计算机视觉与模式识别 · 计算机科学 2019-06-12 Ziang Yan , Yiwen Guo , Changshui Zhang
‹ 上一页 1 2 3 10 下一页 ›