Picture for Phillip Rieger

Phillip Rieger

Technical University Darmstadt

FreqFed: A Frequency Analysis-Based Approach for Mitigating Poisoning Attacks in Federated Learning

Add code
Dec 07, 2023
Viaarxiv icon

FLEDGE: Ledger-based Federated Learning Resilient to Inference and Backdoor Attacks

Add code
Oct 03, 2023
Viaarxiv icon

ARGUS: Context-Based Detection of Stealthy IoT Infiltration Attacks

Add code
Feb 16, 2023
Viaarxiv icon

BayBFed: Bayesian Backdoor Defense for Federated Learning

Add code
Jan 23, 2023
Viaarxiv icon

Close the Gate: Detecting Backdoored Models in Federated Learning based on Client-Side Deep Layer Output Analysis

Add code
Oct 14, 2022
Figure 1 for Close the Gate: Detecting Backdoored Models in Federated Learning based on Client-Side Deep Layer Output Analysis
Figure 2 for Close the Gate: Detecting Backdoored Models in Federated Learning based on Client-Side Deep Layer Output Analysis
Figure 3 for Close the Gate: Detecting Backdoored Models in Federated Learning based on Client-Side Deep Layer Output Analysis
Figure 4 for Close the Gate: Detecting Backdoored Models in Federated Learning based on Client-Side Deep Layer Output Analysis
Viaarxiv icon

DeepSight: Mitigating Backdoor Attacks in Federated Learning Through Deep Model Inspection

Add code
Jan 03, 2022
Figure 1 for DeepSight: Mitigating Backdoor Attacks in Federated Learning Through Deep Model Inspection
Figure 2 for DeepSight: Mitigating Backdoor Attacks in Federated Learning Through Deep Model Inspection
Figure 3 for DeepSight: Mitigating Backdoor Attacks in Federated Learning Through Deep Model Inspection
Figure 4 for DeepSight: Mitigating Backdoor Attacks in Federated Learning Through Deep Model Inspection
Viaarxiv icon