Efficient Adversarial Attack

Efficient adversarial attacks aim to craft minimally perceptible input perturbations that cause machine learning models to misclassify or malfunction, revealing vulnerabilities and driving improvements in model robustness. Current research focuses on developing attacks tailored to specific model architectures, such as graph neural networks and vision-language models, and employing advanced optimization techniques like multi-objective memetic algorithms and gradient-guided methods to enhance attack efficiency and effectiveness. This research is crucial for evaluating and improving the security and reliability of machine learning systems across diverse applications, from image recognition to natural language processing and multi-agent reinforcement learning.

Papers

June 13, 2024

MirrorCheck: Efficient Adversarial Defense for Vision-Language Models
Samar Fares, Klea Ziu, Toluwani Aremu, Nikita Durasov, Martin Takáč, Pascal Fua, Karthik Nandakumar, Ivan Laptev
Adversarial Attack Vision Language Model Adversarial Sample Image Domain Adversarial Threat Efficient Adversarial Attack

March 18, 2024

SSCAE -- Semantic, Syntactic, and Context-aware natural language Adversarial Examples generator
Javad Rafiei Asl, Mohammad H. Rafiei, Manar Alohaly, Daniel Takabi
Language Model Adversarial Attack Adversarial Example Semantic Description Efficient Adversarial Attack Adversarial Context Natural Language Generator

August 16, 2023

Expressivity of Graph Neural Networks Through the Lens of Adversarial Robustness
Francesco Campi, Lukas Gosch, Tom Wollschläger, Yan Scholten, Stephan Günnemann
Graph Neural Network Adversarial Robustness Camera Lens Behavior Expressivity Style Efficient Adversarial Attack

July 15, 2023

Efficient Adversarial Attacks on Online Multi-agent Reinforcement Learning
Guanlin Liu, Lifeng Lai
Adversarial Attack Multi Agent Reinforcement Learning Poisoning Attack Efficient Adversarial Attack

February 24, 2023

HyperAttack: Multi-Gradient-Guided White-box Adversarial Structure Attack of Hypergraph Neural Networks
Chao Hu, Ruishi Yu, Binqi Zeng, Yu Zhan, Ying Fu, Quan Zhang, Rongkai Liu, Heyuan Shi
Adversarial Attack Hypergraph Neural Network White Box Adversarial Attack Efficient Adversarial Attack White Box Adversarial

August 15, 2022

A Multi-objective Memetic Algorithm for Auto Adversarial Attack Optimization Design
Jialiang Sun, Wen Yao, Tingsong Jiang, Xiaoqian Chen
Adversarial Attack Memetic Algorithm Optimal Adversarial Efficient Adversarial Attack

July 25, 2022

SegPGD: An Effective and Efficient Adversarial Attack for Evaluating and Boosting Segmentation Robustness
Jindong Gu, Hengshuang Zhao, Volker Tresp, Philip Torr
Adversarial Example Adversarial Training Adversarial Perturbation Robust Segmentation Efficient Adversarial Attack

June 14, 2022

Adversarial Vulnerability of Randomized Ensembles
Hassan Dbouk, Naresh R. Shanbhag
Adversarial Robustness Adversarial Perturbation Adversarial Vulnerability Ensemble Classifier Imperceptible Adversarial Perturbation Efficient Adversarial Attack

March 7, 2022

$A^{3}D$: A Platform of Searching for Robust Neural Architectures and Efficient Adversarial Attacks
Jialiang Sun, Wen Yao, Tingsong Jiang, Chao Li, Xiaoqian Chen
Adversarial Attack Search Query Adversarial Noise Diverse Platform Robust Neural Architecture Efficient Adversarial Attack

March 2, 2022

Adversarial attacks on neural networks through canonical Riemannian foliations
Eliot Tron, Nicolas Couellan, Stéphane Puechmorel
Neural Network Adversarial Attack Application Proficiency Fisher Information Metric Efficient Adversarial Attack

Efficient Adversarial Attack

Papers

MirrorCheck: Efficient Adversarial Defense for Vision-Language Models

SSCAE -- Semantic, Syntactic, and Context-aware natural language Adversarial Examples generator

Expressivity of Graph Neural Networks Through the Lens of Adversarial Robustness

Efficient Adversarial Attacks on Online Multi-agent Reinforcement Learning

HyperAttack: Multi-Gradient-Guided White-box Adversarial Structure Attack of Hypergraph Neural Networks

A Multi-objective Memetic Algorithm for Auto Adversarial Attack Optimization Design

SegPGD: An Effective and Efficient Adversarial Attack for Evaluating and Boosting Segmentation Robustness

Adversarial Vulnerability of Randomized Ensembles

$A^{3}D$: A Platform of Searching for Robust Neural Architectures and Efficient Adversarial Attacks

Adversarial attacks on neural networks through canonical Riemannian foliations