ImageNet Attack

ImageNet attacks explore vulnerabilities in deep learning models, particularly convolutional neural networks (CNNs), by crafting subtly altered images that cause misclassification. Current research focuses on developing increasingly effective attack methods, targeting various aspects like illumination, latent space manipulation, and even pixel-level perturbations across different $\ell_p$-norms, often using gradient-based or reinforcement learning approaches. These studies highlight the fragility of these models and drive the development of more robust defenses, impacting the reliability and security of AI systems in diverse applications such as image recognition, video processing, and large language models.

Papers

October 6, 2024

TA3: Testing Against Adversarial Attacks on Machine Learning Models
Yuanzhe Jin, Min Chen
Machine Learning Adversarial Attack Machine Learning Model ImageNet Attack

September 5, 2024

Non-Uniform Illumination Attack for Fooling Convolutional Neural Networks
Akshay Jain, Shiv Ram Dubey, Satish Kumar Singh, KC Santosh, Bidyut Baran Chaudhuri
Convolutional Neural Network CNN Performance ImageNet Attack Lightness Adaptation

July 22, 2024

Targeted Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs
Abhay Sheshadri, Aidan Ewart, Phillip Guo, Aengus Lynch, Cindy Wu, Vivek Hebbar, Henry Sleight, Asa Cooper Stickland, Ethan Perez, Dylan Hadfield-Menell, Stephen Casper
Large Language Model Native Robustness Adversarial Training Latent Encoding ImageNet Attack

September 25, 2023

Adversarial Attacks on Video Object Segmentation with Hard Region Discovery
Ping Li, Yu Zhang, Li Yuan, Jian Zhao, Xianghua Xu, Xiaoqin Zhang
Adversarial Attack Adversarial Example Video Object Segmentation Region Selection ImageNet Attack

March 30, 2023

Fooling the Image Dehazing Models by First Order Gradient
Jie Gui, Xiaofeng Cong, Chengwei Peng, Yuan Yan Tang, James Tin-Yau Kwok
Image Dehazing Homogeneous Haze ImageNet Attack First Order Gradient

December 5, 2022

Multiple Perturbation Attack: Attack Pixelwise Under Different $\ell_p$-norms For Better Adversarial Performance
Ngoc N. Tran, Anh Tuan Bui, Dinh Phung, Trung Le
Adversarial Attack Adversarial Example \Ell_p$ Norm Perturbation Attack Better Adversarial Robustness ImageNet Attack

November 15, 2022

Universal Distributional Decision-based Black-box Adversarial Attack with Reinforcement Learning
Yiran Huang, Yexu Zhou, Michael Hefenbrock, Till Riedel, Likun Fang, Michael Beigl
Reinforcement Learning Adversarial Attack Adversarial Perturbation Black Box Adversarial Attack Decision Based ImageNet Attack

April 30, 2022

Optimizing One-pixel Black-box Adversarial Attacks
Tianxun Zhou, Shubhankar Agrawal, Prateek Manocha
Deep Neural Network Adversarial Attack Pixel Wise Black Box Adversarial Attack Image Perturbation ImageNet Attack

January 27, 2022

Beyond ImageNet Attack: Towards Crafting Adversarial Examples for Black-box Domains
Qilong Zhang, Xiaodan Li, Yuefeng Chen, Jingkuan Song, Lianli Gao, Yuan He, Hui Xue
Adversarial Example Black Box Function Model Transferability ImageNet Attack