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

March 15, 2024

Towards Non-Adversarial Algorithmic Recourse
Tobias Leemann, Martin Pawelczyk, Bardh Prenkaj, Gjergji Kasneci
Adversarial Example

March 14, 2024

Counter-Samples: A Stateless Strategy to Neutralize Black Box Adversarial Attacks
Roey Bokobza, Yisroel Mirsky
Adversarial Example Black Box Attack Defense Method Black Box Adversarial Attack Unlabeled Client

March 12, 2024

Disentangling Policy from Offline Task Representation Learning via Adversarial Data Augmentation
Chengxing Jia, Fuxiang Zhang, Yi-Chen Li, Chen-Xiao Gao, Xu-Hui Liu, Lei Yuan, Zongzhang Zhang, Yang Yu
Contrastive Learning Adversarial Example Health Policy Recommendation Task Specification Task Representation Adversarial Data Augmentation Offline Meta Reinforcement Learning Offline Visual Representation Learning

March 11, 2024

March 8, 2024

Adversarial Sparse Teacher: Defense Against Distillation-Based Model Stealing Attacks Using Adversarial Examples
Eda Yilmaz, Hacer Yalim Keles
Knowledge Distillation Adversarial Example Adversarial Training Discriminative Power Model Stealing Attack

March 6, 2024

Improving Adversarial Training using Vulnerability-Aware Perturbation Budget
Olukorede Fakorede, Modeste Atsague, Jin Tian
Adversarial Attack Adversarial Example Adversarial Training Adversarial Sample

March 5, 2024

Towards Robust Federated Learning via Logits Calibration on Non-IID Data
Yu Qiao, Apurba Adhikary, Chaoning Zhang, Choong Seon Hong
Adversarial Attack Adversarial Example Adversarial Training Raw Data Second Ranked Logits

March 4, 2024

February 29, 2024

February 28, 2024

Catastrophic Overfitting: A Potential Blessing in Disguise
Mengnan Zhao, Lihe Zhang, Yuqiu Kong, Baocai Yin
Adversarial Example Adversarial Robustness Adversarial Data Digital STEALTH Metric Fast Adversarial Training Catastrophic Overfitting Blessing Generation

February 27, 2024

Adversarial Example

Papers

Towards Non-Adversarial Algorithmic Recourse

Counter-Samples: A Stateless Strategy to Neutralize Black Box Adversarial Attacks

Disentangling Policy from Offline Task Representation Learning via Adversarial Data Augmentation

Dynamic Perturbation-Adaptive Adversarial Training on Medical Image Classification

PeerAiD: Improving Adversarial Distillation from a Specialized Peer Tutor

Adversarial Sparse Teacher: Defense Against Distillation-Based Model Stealing Attacks Using Adversarial Examples

Improving Adversarial Training using Vulnerability-Aware Perturbation Budget

Towards Robust Federated Learning via Logits Calibration on Non-IID Data

Robustness Bounds on the Successful Adversarial Examples: Theory and Practice

One Prompt Word is Enough to Boost Adversarial Robustness for Pre-trained Vision-Language Models

Unraveling Adversarial Examples against Speaker Identification -- Techniques for Attack Detection and Victim Model Classification

Pointing out the Shortcomings of Relation Extraction Models with Semantically Motivated Adversarials

How to Train your Antivirus: RL-based Hardening through the Problem-Space

MPAT: Building Robust Deep Neural Networks against Textual Adversarial Attacks

Enhancing the "Immunity" of Mixture-of-Experts Networks for Adversarial Defense

Catastrophic Overfitting: A Potential Blessing in Disguise

Adversarial Math Word Problem Generation

Black-box Adversarial Attacks Against Image Quality Assessment Models

Extreme Miscalibration and the Illusion of Adversarial Robustness

Robustness-Congruent Adversarial Training for Secure Machine Learning Model Updates