Resilient Backdoor

Resilient backdoor attacks aim to embed malicious functionality into machine learning models, making them vulnerable to manipulation by injecting specific triggers that cause misclassification. Current research focuses on developing more robust backdoor techniques that resist detection and mitigation methods, including those leveraging continual learning, architectural modifications, and sophisticated trigger generation strategies designed to evade detection. This area is crucial because the persistence and difficulty of detecting these backdoors pose significant security risks to various machine learning applications, demanding the development of more effective defenses.

Papers

July 20, 2024

Flatness-aware Sequential Learning Generates Resilient Backdoors
Hoang Pham, The-Anh Ta, Anh Tran, Khoa D. Doan
Fine Tuning Backdoor Attack Sequential Learning Backdoor Learning Input Level Backdoor Detection Resilient Backdoor

May 30, 2024

Unveiling and Mitigating Backdoor Vulnerabilities based on Unlearning Weight Changes and Backdoor Activeness
Weilin Lin, Li Liu, Shaokui Wei, Jianze Li, Hui Xiong
Backdoor Defense Weight Freezing Neural Backdoor Backdoor Effect Resilient Backdoor

March 25, 2024

LOTUS: Evasive and Resilient Backdoor Attacks through Sub-Partitioning
Siyuan Cheng, Guanhong Tao, Yingqi Liu, Guangyu Shen, Shengwei An, Shiwei Feng, Xiangzhe Xu, Kaiyuan Zhang, Shiqing Ma, Xiangyu Zhang
Backdoor Attack Backdoor Detection Backdoor Behavior Resilient Backdoor Lung Originated Tumor Segmentation

February 10, 2024

Architectural Neural Backdoors from First Principles
Harry Langford, Ilia Shumailov, Yiren Zhao, Robert Mullins, Nicolas Papernot
Backdoor Attack First Principle Neural Backdoor Backdoor Behavior Resilient Backdoor

November 28, 2023

Rethinking Backdoor Attacks on Dataset Distillation: A Kernel Method Perspective
Ming-Yu Chung, Sheng-Yen Chou, Chia-Mu Yu, Pin-Yu Chen, Sy-Yen Kuo, Tsung-Yi Ho
Backdoor Attack Dataset Distillation Kernel Method Effective Backdoor Attack Trigger Design Resilient Backdoor

April 25, 2023

Chameleon: Adapting to Peer Images for Planting Durable Backdoors in Federated Learning
Yanbo Dai, Songze Li
Backdoor Attack Peer Agent Pace Adaptive Malicious Client Resilient Backdoor

April 20, 2023

Get Rid Of Your Trail: Remotely Erasing Backdoors in Federated Learning
Manaar Alam, Hithem Lamri, Michail Maniatakos
Backdoor Attack Root to Leaf Path Active Removal Backdoor Removal Resilient Backdoor

March 1, 2023

Single Image Backdoor Inversion via Robust Smoothed Classifiers
Mingjie Sun, J. Zico Kolter
Smoothed Classifier Effective Backdoor Attack Resilient Backdoor Backdoor Inversion

June 15, 2022

Architectural Backdoors in Neural Networks
Mikel Bober-Irizar, Ilia Shumailov, Yiren Zhao, Robert Mullins, Nicolas Papernot
Neural Network Backdoor Attack Inductive Bias Adversarial Manipulation Resilient Backdoor

June 12, 2022

Neurotoxin: Durable Backdoors in Federated Learning
Zhengming Zhang, Ashwinee Panda, Linyue Song, Yaoqing Yang, Michael W. Mahoney, Joseph E. Gonzalez, Kannan Ramchandran, Prateek Mittal
Backdoor Attack Adversarial Backdoor Resilient Backdoor

February 19, 2022

Label-Smoothed Backdoor Attack
Minlong Peng, Zidi Xiong, Mingming Sun, Ping Li
Backdoor Attack Attack Success Rate Clean Label Backdoor Attack Resilient Backdoor