NOTE(More information)

CerberusFL: Secure and Decentralized Federated Learning with AES Encryption, Weighted Averaging Aggregation, and Malicious Participant Detection

CerberusFL is a new blockchain algorithm that enables secure and decentralized federated learning. Federated learning is a machine learning technique that allows multiple participants to train a shared model without sharing their data. CerberusFL uses blockchain to ensure the security and privacy of the federated learning process.

Features:

AES encryption for secure global model updates
Weighted averaging aggregation for more accurate and robust global models
Malicious participant detection to prevent malicious participants from degrading the quality of the global model or compromising the privacy of the participants' data
Benefits:

Increased security
Improved accuracy
Increased robustness
Enhanced privacy
Potential Applications:

Healthcare
Finance
Cybersecurity
Cloud computing