Federated Learning is a data privacy-preserving approach that allows machine learning models with decentralized data samples without exchanging raw data. Conventional Machine Learning approaches on malware detection in IoT network based on the entire data located on a central server is significantly challenging for domains with privacy concerns on user data.
   Client access and limited network reliability are significant research challenges in developing a federated deep learning system for malware detection in IoT networks. To tackle this constraint, Deep Federated Learning models adopted supervised and unsupervised models for proactively detecting the malware in IoT networks without sharing sensitive data and using decentralized on-device data.