Space Attack

Space attack research focuses on developing and defending against adversarial attacks targeting machine learning models, particularly in image recognition and autonomous systems. Current efforts concentrate on crafting increasingly sophisticated attacks, such as those manipulating feature spaces subtly (e.g., through seemingly innocuous image alterations) or exploiting vulnerabilities in graph-based anomaly detection systems, often employing optimization techniques and Markov Decision Processes to model attacker behavior and defense strategies. This research is crucial for enhancing the security and reliability of AI systems across various applications, from facial recognition to autonomous driving, by identifying weaknesses and developing robust countermeasures.

Papers

August 22, 2024

MakeupAttack: Feature Space Black-box Backdoor Attack on Face Recognition via Makeup Transfer
Ming Sun, Lihua Jing, Zixuan Zhu, Rui Wang
Face Recognition Makeup Product Image Facial Datasets Makeup Transfer Adversarial Backdoor Space Attack

August 16, 2024

A Factored MDP Approach To Moving Target Defense With Dynamic Threat Modeling and Cost Efficiency
Megha Bose, Praveen Paruchuri, Akshat Kumar
Markov Decision Process Threat Model Cost Effective Moving Target Defense Space Attack

July 26, 2023

Coupled-Space Attacks against Random-Walk-based Anomaly Detection
Yuni Lai, Marcin Waniek, Liying Li, Jingwen Wu, Yulin Zhu, Tomasz P. Michalak, Talal Rahwan, Kai Zhou
Anomaly Detection New Attack Graph Anomaly Detection Attack Graph Space Attack

October 7, 2022

A2: Efficient Automated Attacker for Boosting Adversarial Training
Zhuoer Xu, Guanghui Zhu, Changhua Meng, Shiwen Cui, Zhenzhe Ying, Weiqiang Wang, Ming GU, Yihua Huang
Native Robustness Adversarial Training Model Robustness Robust Model Space Attack

July 20, 2022

Illusory Attacks: Information-Theoretic Detectability Matters in Adversarial Attacks
Tim Franzmeyer, Stephen McAleer, João F. Henriques, Jakob N. Foerster, Philip H. S. Torr, Adel Bibi, Christian Schroeder de Witt
Adversarial Attack Information Theoretic Detectability Imperceptible Attack Space Attack

May 18, 2022

Defending Object Detectors against Patch Attacks with Out-of-Distribution Smoothing
Ryan Feng, Neal Mangaokar, Jihye Choi, Somesh Jha, Atul Prakash
Native Robustness Machine Learning Model Outlier Detection Car Learning to Act Preprocessing Method Patch Attack Space Attack

Space Attack

Papers

MakeupAttack: Feature Space Black-box Backdoor Attack on Face Recognition via Makeup Transfer

A Factored MDP Approach To Moving Target Defense With Dynamic Threat Modeling and Cost Efficiency

Coupled-Space Attacks against Random-Walk-based Anomaly Detection

A2: Efficient Automated Attacker for Boosting Adversarial Training

Illusory Attacks: Information-Theoretic Detectability Matters in Adversarial Attacks

Defending Object Detectors against Patch Attacks with Out-of-Distribution Smoothing