Model Poisoning Attack

Model poisoning attacks target federated learning (FL) systems by injecting malicious model updates from compromised clients, aiming to degrade the global model's accuracy or introduce biases. Current research focuses on developing robust aggregation techniques, such as those employing uncertainty-aware evaluation, dynamic weighting, or Fourier transforms, to identify and mitigate the effects of these poisoned updates, often in conjunction with client-side defenses. Understanding and defending against these attacks is crucial for ensuring the reliability and trustworthiness of FL, a technology with significant potential for privacy-preserving machine learning across diverse applications.

Papers

December 5, 2022

FedCC: Robust Federated Learning against Model Poisoning Attacks
Hyejun Jeong, Hamin Son, Seohu Lee, Jayun Hyun, Tai-Myoung Chung
Model Poisoning Attack Malicious Client

December 4, 2022

Security Analysis of SplitFed Learning
Momin Ahmad Khan, Virat Shejwalkar, Amir Houmansadr, Fatima Muhammad Anwar
Split Learning Model Poisoning Attack Splitfed Learning

November 7, 2022

Resilience of Wireless Ad Hoc Federated Learning against Model Poisoning Attacks
Naoya Tezuka, Hideya Ochiai, Yuwei Sun, Hiroshi Esaki
Federated Learning Tiny Refinement Elicit Resilience Collaborative Machine Learning Model Poisoning Attack Strong Model Wireless Node

October 17, 2022

Thinking Two Moves Ahead: Anticipating Other Users Improves Backdoor Attacks in Federated Learning
Yuxin Wen, Jonas Geiping, Liam Fowl, Hossein Souri, Rama Chellappa, Micah Goldblum, Tom Goldstein
Backdoor Attack New Attack Model Poisoning Attack Effective Non Local Move

August 16, 2022

FedPerm: Private and Robust Federated Learning by Parameter Permutation
Hamid Mozaffari, Virendra J. Marathe, Dave Dice
Differential Privacy Secure Aggregation Adversarial Behavior Model Poisoning Attack

July 19, 2022

FLDetector: Defending Federated Learning Against Model Poisoning Attacks via Detecting Malicious Clients
Zaixi Zhang, Xiaoyu Cao, Jinyuan Jia, Neil Zhenqiang Gong
Model Poisoning Attack Byzantine Robust Federated Learning Malicious Client

April 29, 2022

Backdoor Attacks in Federated Learning by Rare Embeddings and Gradient Ensembling
KiYoon Yoo, Nojun Kwak
Backdoor Attack Word Embeddings Model Ensembling Model Poisoning Attack Adversarial Objective Backdoor Behavior

April 22, 2022

Federated Learning: Balancing the Thin Line Between Data Intelligence and Privacy
Sherin Mary Mathews, Samuel A. Assefa
Privacy Policy Inference Attack Model Poisoning Attack Security Challenge Business Intelligence Fragmented Personal Data

April 1, 2022

FedRecAttack: Model Poisoning Attack to Federated Recommendation
Dazhong Rong, Shuai Ye, Ruoyan Zhao, Hon Ning Yuen, Jianhai Chen, Qinming He
Recommender System Model Poisoning Attack Malicious User Federated Recommendation

March 22, 2022

Semi-Targeted Model Poisoning Attack on Federated Learning via Backward Error Analysis
Yuwei Sun, Hideya Ochiai, Jun Sakuma
Poisoning Attack Model Poisoning Attack Attack Performance Based Attack

March 16, 2022

MPAF: Model Poisoning Attacks to Federated Learning based on Fake Clients
Xiaoyu Cao, Neil Zhenqiang Gong
Model Poisoning Attack Fake User Local Update Attacker Model

February 6, 2022

BEAS: Blockchain Enabled Asynchronous & Secure Federated Machine Learning
Arup Mondal, Harpreet Virk, Debayan Gupta
Federated Learning Blockchain Based Platform Differential Privacy Guarantee Private Datasets Model Poisoning Attack Data Reconstruction Attack Asynchronous Architecture

December 12, 2021

SparseFed: Mitigating Model Poisoning Attacks in Federated Learning with Sparsification
Ashwinee Panda, Saeed Mahloujifar, Arjun N. Bhagoji, Supriyo Chakraborty, Prateek Mittal
Poisoning Attack Random Sparsification Model Poisoning Attack Top K Sparsification

Model Poisoning Attack

Papers

FedCC: Robust Federated Learning against Model Poisoning Attacks

Security Analysis of SplitFed Learning

Resilience of Wireless Ad Hoc Federated Learning against Model Poisoning Attacks

Thinking Two Moves Ahead: Anticipating Other Users Improves Backdoor Attacks in Federated Learning

FedPerm: Private and Robust Federated Learning by Parameter Permutation

FLDetector: Defending Federated Learning Against Model Poisoning Attacks via Detecting Malicious Clients

Backdoor Attacks in Federated Learning by Rare Embeddings and Gradient Ensembling

Federated Learning: Balancing the Thin Line Between Data Intelligence and Privacy

FedRecAttack: Model Poisoning Attack to Federated Recommendation

Semi-Targeted Model Poisoning Attack on Federated Learning via Backward Error Analysis

MPAF: Model Poisoning Attacks to Federated Learning based on Fake Clients

BEAS: Blockchain Enabled Asynchronous & Secure Federated Machine Learning

SparseFed: Mitigating Model Poisoning Attacks in Federated Learning with Sparsification