Adversarial Example

Adversarial examples are subtly altered inputs designed to fool machine learning models, primarily deep neural networks (DNNs), into making incorrect predictions. Current research focuses on improving model robustness against these attacks, exploring techniques like ensemble methods, multi-objective representation learning, and adversarial training, often applied to architectures such as ResNets and Vision Transformers. Understanding and mitigating the threat of adversarial examples is crucial for ensuring the reliability and security of AI systems across diverse applications, from image classification and natural language processing to malware detection and autonomous driving. The development of robust defenses and effective attack detection methods remains a significant area of ongoing investigation.

Papers

April 20, 2022

Adversarial Scratches: Deployable Attacks to CNN Classifiers
Loris Giulivi, Malhar Jere, Loris Rossi, Farinaz Koushanfar, Gabriela Ciocarlie, Briland Hitaj, Giacomo Boracchi
Adversarial Example Black Box Attack CNN Classifier Adversarial Disturbance

April 19, 2022

Generating Authentic Adversarial Examples beyond Meaning-preserving with Doubly Round-trip Translation
Siyu Lai, Zhen Yang, Fandong Meng, Xue Zhang, Yufeng Chen, Jinan Xu, Jie Zhou
Adversarial Example Neural Machine Translation Neural Machine Translation Model Round Trip Adversarial Pair

April 18, 2022

April 17, 2022

Residue-Based Natural Language Adversarial Attack Detection
Vyas Raina, Mark Gales
Adversarial Attack Adversarial Example Adversarial Detection Language Based

April 14, 2022

Planting Undetectable Backdoors in Machine Learning Models
Shafi Goldwasser, Michael P. Kim, Vinod Vaikuntanathan, Or Zamir
Machine Learning Model Adversarial Example Adversarial Robustness

April 13, 2022

April 12, 2022

Examining the Proximity of Adversarial Examples to Class Manifolds in Deep Networks
Štefan Pócoš, Iveta Bečková, Igor Farkaš
Deep Neural Network Adversarial Example Deep Network Internal Representation Adversarial Text Class Manifold Hidden Layer Proximity Search

April 11, 2022

A Simple Approach to Adversarial Robustness in Few-shot Image Classification
Akshayvarun Subramanya, Hamed Pirsiavash
Adversarial Example Adversarial Robustness Shot Learning Shot Image Classification Simple Approach Shot Classifier ROBust Few Shot

April 8, 2022

April 7, 2022

Adaptive-Gravity: A Defense Against Adversarial Samples
Ali Mirzaeian, Zhi Tian, Sai Manoj P D, Banafsheh S. Latibari, Ioannis Savidis, Houman Homayoun, Avesta Sasan
Adversarial Attack Adversarial Example Adversarial Sample Gravity Compensation

April 6, 2022

April 3, 2022

DST: Dynamic Substitute Training for Data-free Black-box Attack
Wenxuan Wang, Xuelin Qian, Yanwei Fu, Xiangyang Xue
Adversarial Example Substitute Model Dynamic Training Synonym Substitution Attack DST Det Data Free Black Box Attack

March 31, 2022

Adversarial Examples in Random Neural Networks with General Activations
Andrea Montanari, Yuchen Wu
Deep Learning Model Adversarial Example Targeted Activation Penalty Layer ReLU Network Random Neural Network Two Layer Network

March 30, 2022

Sensor Data Validation and Driving Safety in Autonomous Driving Systems
Jindi Zhang
Autonomous Driving Autonomous Vehicle Adversarial Example High Precision Lidar Sensory Data

March 28, 2022

Boosting Black-Box Adversarial Attacks with Meta Learning
Junjie Fu, Jian Sun, Gang Wang
Adversarial Attack Adversarial Example Black Box Adversarial Attack

March 26, 2022

A Survey of Robust Adversarial Training in Pattern Recognition: Fundamental, Theory, and Methodologies
Zhuang Qian, Kaizhu Huang, Qiu-Feng Wang, Xu-Yao Zhang
Timely Survey Adversarial Attack Adversarial Example Adversarial Training NCD Method Theoretical Understanding Adversarial Sample Pattern Recognition Robust Adversarial Training

Adversarial Example

Papers

Adversarial Scratches: Deployable Attacks to CNN Classifiers

Generating Authentic Adversarial Examples beyond Meaning-preserving with Doubly Round-trip Translation

UNBUS: Uncertainty-aware Deep Botnet Detection System in Presence of Perturbed Samples

Sardino: Ultra-Fast Dynamic Ensemble for Secure Visual Sensing at Mobile Edge

Residue-Based Natural Language Adversarial Attack Detection

Planting Undetectable Backdoors in Machine Learning Models

A Novel Approach to Train Diverse Types of Language Models for Health Mention Classification of Tweets

Synthesizing Adversarial Visual Scenarios for Model-Based Robotic Control

Examining the Proximity of Adversarial Examples to Class Manifolds in Deep Networks

A Simple Approach to Adversarial Robustness in Few-shot Image Classification

Evaluating the Adversarial Robustness for Fourier Neural Operators

AdvEst: Adversarial Perturbation Estimation to Classify and Detect Adversarial Attacks against Speaker Identification

Adaptive-Gravity: A Defense Against Adversarial Samples

Sampling-based Fast Gradient Rescaling Method for Highly Transferable Adversarial Attacks

Distilling Robust and Non-Robust Features in Adversarial Examples by Information Bottleneck

DST: Dynamic Substitute Training for Data-free Black-box Attack

Adversarial Examples in Random Neural Networks with General Activations

Sensor Data Validation and Driving Safety in Autonomous Driving Systems

Boosting Black-Box Adversarial Attacks with Meta Learning

A Survey of Robust Adversarial Training in Pattern Recognition: Fundamental, Theory, and Methodologies