Backdoor Poisoning Attack

Backdoor poisoning attacks exploit vulnerabilities in machine learning models by injecting malicious data into training datasets. Current research focuses on developing more sophisticated attacks, particularly clean-label attacks that avoid obvious label inconsistencies, and exploring various trigger mechanisms, including physical triggers and transformations like rotation, across diverse model architectures such as deep neural networks, graph neural networks, and reinforcement learning agents. The significance of this research lies in its implications for the security and trustworthiness of machine learning systems across numerous applications, from image recognition and speech processing to critical infrastructure control, necessitating the development of robust defenses.

Papers

October 17, 2024

Adversarial Inception for Bounded Backdoor Poisoning in Deep Reinforcement Learning
Ethan Rathbun, Christopher Amato, Alina Oprea
Reinforcement Learning Backdoor Attack Adversarial Learning Deep Reinforcement Adversarial Behavior Backdoor Poisoning Attack

September 13, 2024

Clean Label Attacks against SLU Systems
Henry Li Xinyuan, Sonal Joshi, Thomas Thebaud, Jesus Villalba, Najim Dehak, Sanjeev Khudanpur
Data Poisoning Attack Gradient Based Attack Backdoor Poisoning Attack End 2 End Clean Label Backdoor Attack Clean Label Attack

July 27, 2024

Towards Clean-Label Backdoor Attacks in the Physical World
Thinh Dao, Cuong Chi Le, Khoa D Doan, Kok-Seng Wong
Backdoor Poisoning Attack Physical World Clean Label Backdoor Attack Physical Backdoor Attack Physical Attack

June 23, 2024

CBPF: Filtering Poisoned Data Based on Composite Backdoor Attack
Hanfeng Xia, Haibo Hong, Ruili Wang
Backdoor Attack Backdoor Poisoning Attack Poisoned Data Composite Attack Target Class

May 30, 2024

SleeperNets: Universal Backdoor Poisoning Attacks Against Reinforcement Learning Agents
Ethan Rathbun, Christopher Amato, Alina Oprea
Reinforcement Learning Adversarial Attack Reinforcement Learning Agent Backdoor Poisoning Attack Training Time Attack Universal Backdoor

May 25, 2024

Mitigating Backdoor Attack by Injecting Proactive Defensive Backdoor
Shaokui Wei, Hongyuan Zha, Baoyuan Wu
Backdoor Defense Backdoor Poisoning Attack Mitigating Backdoor

May 19, 2024

SEEP: Training Dynamics Grounds Latent Representation Search for Mitigating Backdoor Poisoning Attacks
Xuanli He, Qiongkai Xu, Jun Wang, Benjamin I. P. Rubinstein, Trevor Cohn
Backdoor Attack Latent Representation Data Poisoning Attack Backdoor Poisoning Attack

May 17, 2024

Rethinking Graph Backdoor Attacks: A Distribution-Preserving Perspective
Zhiwei Zhang, Minhua Lin, Enyan Dai, Suhang Wang
Graph Neural Network Backdoor Attack Backdoor Trigger Backdoor Poisoning Attack Graph Backdoor Attack

May 9, 2024

Poisoning-based Backdoor Attacks for Arbitrary Target Label with Positive Triggers
Binxiao Huang, Jason Chun Lok, Chang Liu, Ngai Wong
Backdoor Attack Backdoor Defense Adversarial Method Backdoor Poisoning Attack Trigger Design Target Class Trigger Warning

March 15, 2024

Backdoor Secrets Unveiled: Identifying Backdoor Data with Optimized Scaled Prediction Consistency
Soumyadeep Pal, Yuguang Yao, Ren Wang, Bingquan Shen, Sijia Liu
Backdoor Poisoning Attack Consistent Prediction Backdoor Data

February 19, 2024

Defending Against Weight-Poisoning Backdoor Attacks for Parameter-Efficient Fine-Tuning
Shuai Zhao, Leilei Gan, Luu Anh Tuan, Jie Fu, Lingjuan Lyu, Meihuizi Jia, Jinming Wen
Backdoor Attack Parameter Efficient Fine Tuning Backdoor Poisoning Attack Robust Defense Poison Detection

December 19, 2023

BadRL: Sparse Targeted Backdoor Attack Against Reinforcement Learning
Jing Cui, Yufei Han, Yuzhe Ma, Jianbin Jiao, Junge Zhang
Reinforcement Learning Backdoor Attack Attack Strategy Backdoor Poisoning Attack

December 18, 2023

DataElixir: Purifying Poisoned Dataset to Mitigate Backdoor Attacks via Diffusion Models
Jiachen Zhou, Peizhuo Lv, Yibing Lan, Guozhu Meng, Kai Chen, Hualong Ma
Diffusion Model Data Set Backdoor Attack Backdoor Poisoning Attack Clean Data Cut and Paste

October 16, 2023

Demystifying Poisoning Backdoor Attacks from a Statistical Perspective
Ganghua Wang, Xun Xian, Jayanth Srinivasa, Ashish Kundu, Xuan Bi, Mingyi Hong, Jie Ding
Backdoor Attack Backdoor Poisoning Attack Effective Backdoor Attack Backdoor Data

September 28, 2023

Post-Training Overfitting Mitigation in DNN Classifiers
Hang Wang, David J. Miller, George Kesidis
Adversarial Training Adversarial Learning Model Overfitting Benign Overfitting Backdoor Poisoning Attack Deep Neural Network Classifier

July 17, 2023

Towards Stealthy Backdoor Attacks against Speech Recognition via Elements of Sound
Hanbo Cai, Pengcheng Zhang, Hai Dong, Yan Xiao, Stefanos Koffas, Yiming Li
Speech Recognition Backdoor Attack Sound Design Backdoor Poisoning Attack Element Level Information Stealthy Backdoor Attack Imperceptible Trigger Inaudible Voice

June 14, 2023

A Proxy Attack-Free Strategy for Practically Improving the Poisoning Efficiency in Backdoor Attacks
Ziqiang Li, Hong Sun, Pengfei Xia, Beihao Xia, Xue Rui, Wei Zhang, Qinglang Guo, Bin Li
Backdoor Attack Backdoor Poisoning Attack Effective Backdoor Attack Weapon of Mass Destruction Proxy Poisoning Efficiency

May 28, 2023

Backdoor Attacks Against Incremental Learners: An Empirical Evaluation Study
Yiqi Zhong, Xianming Liu, Deming Zhai, Junjun Jiang, Xiangyang Ji
Backdoor Attack Empirical Evaluation Backdoor Poisoning Attack Task Attack

May 19, 2023

Mitigating Backdoor Poisoning Attacks through the Lens of Spurious Correlation
Xuanli He, Qiongkai Xu, Jun Wang, Benjamin Rubinstein, Trevor Cohn
Backdoor Attack Camera Lens Malicious Agent Backdoor Poisoning Attack Insertion Based

January 27, 2023

PECAN: A Deterministic Certified Defense Against Backdoor Attacks
Yuhao Zhang, Aws Albarghouthi, Loris D'Antoni
Backdoor Attack Backdoor Poisoning Attack Content Addressable Memory Certified Defense Test Time Defense