Adversarial Data

Adversarial data, encompassing maliciously perturbed inputs designed to mislead machine learning models, poses a significant threat to the reliability of AI systems. Current research focuses on developing robust models through techniques like adversarial training (incorporating adversarial examples during training), and novel detection methods that identify adversarial instances based on distributional discrepancies or feature analysis, often employing diffusion models or ensemble approaches. This field is crucial for ensuring the trustworthiness and security of AI applications across diverse domains, from medical diagnosis and autonomous driving to natural language processing, where adversarial attacks can have serious consequences.

Papers

November 27, 2022

Adversarial Rademacher Complexity of Deep Neural Networks
Jiancong Xiao, Yanbo Fan, Ruoyu Sun, Zhi-Quan Luo
Deep Neural Network Adversarial Attack Adversarial Loss Adversarial Data Rademacher Complexity

August 17, 2022

An Evolutionary, Gradient-Free, Query-Efficient, Black-Box Algorithm for Generating Adversarial Instances in Deep Networks
Raz Lapid, Zvika Haramaty, Moshe Sipper
Adversarial Example Deep Network Black Box Adversarial Data QuEry Based Attack Adversarial Instance

July 29, 2022

Robust Trajectory Prediction against Adversarial Attacks
Yulong Cao, Danfei Xu, Xinshuo Weng, Zhuoqing Mao, Anima Anandkumar, Chaowei Xiao, Marco Pavone
Adversarial Attack Autonomous Driving Adversarial Example Trajectory Prediction Adversarial Data

June 28, 2022

Collecting high-quality adversarial data for machine reading comprehension tasks with humans and models in the loop
Damian Y. Romero Diaz, Magdalena Anioł, John Culnan
Full Model Real Human High Quality Machine Reading Comprehension Adversarial Data Adversarial Data Collection Extractive Question Annotation Study Successful Adversarial Attack

June 17, 2022

Understanding Robust Overfitting of Adversarial Training and Beyond
Chaojian Yu, Bo Han, Li Shen, Jun Yu, Chen Gong, Mingming Gong, Tongliang Liu
Adversarial Training Adversarial Robustness Adversarial Data Robust Overfitting Large Loss

June 8, 2022

Adversarial Text Normalization
Joanna Bitton, Maya Pavlova, Ivan Evtimov
Adversarial Attack Adversarial Data Text Normalization Adversarial Retraining

June 2, 2022

Adversarial RAW: Image-Scaling Attack Against Imaging Pipeline
Junjian Li, Honglong Chen
Adversarial Attack Adversarial Data Imaging Pipeline

May 25, 2022

Less Learn Shortcut: Analyzing and Mitigating Learning of Spurious Feature-Label Correlation
Yanrui Du, Jing Yan, Yan Chen, Jing Liu, Sendong Zhao, Qiaoqiao She, Hua Wu, Haifeng Wang, Bing Qin
Dataset Bias Shortcut Learning Adversarial Data Biased Behavior Common Bias

May 9, 2022

Robustness of Humans and Machines on Object Recognition with Extreme Image Transformations
Dakarai Crowder, Girik Malik
Native Robustness New Machine Real Human Deep Learning Architecture Object Recognition Visual Task Image Transformation Visual Cortex Adversarial Data

May 5, 2022

Can collaborative learning be private, robust and scalable?
Dmitrii Usynin, Helena Klause, Johannes C. Paetzold, Daniel Rueckert, Georgios Kaissis
Differential Privacy Adversarial Training Adversarial Sample Collaborative Learning Adversarial Data Robust Adversarial

April 28, 2022

Improving robustness of language models from a geometry-aware perspective
Bin Zhu, Zhaoquan Gu, Le Wang, Jinyin Chen, Qi Xuan
Language Model Native Robustness Adversarial Training Adversarial Data Enhanced Robustness Adversarial Style Augmentation Adversarial Transformation

March 16, 2022

What Do Adversarially trained Neural Networks Focus: A Fourier Domain-based Study
Binxiao Huang, Chaofan Tao, Rui Lin, Ngai Wong
Native Robustness Adversarial Perturbation Adversarial Learning Cross Domain Fourier Transform Adversarial Data Robust Neural Network

February 28, 2022

Markov Chain Monte Carlo-Based Machine Unlearning: Unlearning What Needs to be Forgotten
Quoc Phong Nguyen, Ryutaro Oikawa, Dinil Mon Divakaran, Mun Choon Chan, Bryan Kian Hsiang Low
Machine Learning Training Data Machine Learning Model Machine Unlearning Markov Chain Monte Carlo Model Prediction Adversarial Data

January 3, 2022

Actor-Critic Network for Q&A in an Adversarial Environment
Bejan Sadeghian
Reinforcement Learning Adversarial Attack Actor Critic Adversarial Agent Adversarial Data Adversarial Environment Critic Model

December 23, 2021

Adaptive Modeling Against Adversarial Attacks
Zhiwen Yan, Teck Khim Ng
Adversarial Attack Adversarial Training Adversarial Input Adversarial Data Adversarial Defense Adaptive Model Gradient Attack