Deep Neural Network (DNN) classifiers are vulnerable to adversarial attack, where an imperceptible perturbation could result in misclassification. However, the vulnerability of DNN-based image ranking systems remains under-explored. In this paper, we propose two attacks against deep ranking systems, i....
Adversarial example Adversarial attack Adversarial defense 1. Introduction A trillion-fold increase in computation power has popularized the usage of deep learning (DL) for handling a variety of machine learning (ML) tasks, such as image classification [1], natural language processing [2], and game...
Definition of Adversarial ranking attack: adversarial ranking attack aims raise or lower the ranks of some chosen candidates C={c₁,c₂, ... ,cₘ} with respect to a specific query set Q={q₁,q₂, ... ,qw}. This can be achieved by either Candidate Attack (CA) or Query ...
Multi-granularity Textual Adversarial Attack with Behavior Cloning. Yangyi Chen, Jin Su, Wei Wei. EMNLP 2021. blind [pdf] [code] Synthesizing Adversarial Negative Responses for Robust Response Ranking and Evaluation. Prakhar Gupta, Yulia Tsvetkov, Jeffrey Bigham. Findings of ACL: ACL-IJCNLP 2021....
We propose a novel and low-cost test-time adversarial defense by devising interpretability-guided neuron importance ranking methods to identify neurons important to the output classes. Our method is a training-free approach that can significantly improve the robustness-accuracy tradeoff while incurring ...
H. (2020). Adversarial metric attack and defense for person re-identification. IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(6), 2119–2126. Article Google Scholar Bai, X., Yang, M., Huang, T., Dou, Z., Yu, R., & Xu, Y. (2020). Deep-person: Learning ...
Evaluation processes include systematic ranking, Sensitivity Analysis, and Explainability Analysis, employing the LIME algorithm. These diverse elements collectively contribute to the distinctiveness of the study in the context of VANETs and adversarial attack detection. The structure of the paper is as ...
Christian Szegedy et al. firstly found that the highly accurate modern deep learning models are susceptible to adversarial samples that are derived from the original images by adding small perturbations. To address the security problem of AI system, many methods for model attack and defense have bee...
2020Adversarial Training-Based Mean Bayesian Personalized Ranking for Recommender SystemAttack/DefenseBPR-MFIEEE AccessLink 2020Adversarial Learning to Compare: Self-Attentive Prospective Customer Recommendation in Location based Social NetworksAttack/DefenseLBSNWSDMLink ...
Finally, combined with the current status of adversarial example generation and defense technology development, put forward challenges and prospects in this field. Keywords: adversarial example; deep neural network; smart city; adversarial defense; black-box attack; white-box attack...