1. INTRODUCTION
Federated learning has recently emerged as a major tenet of machine learning and deep learning due to growing privacy concerns associated with the large amounts of user data that these models depend on. Such information should not be accessible to third parties, who, upon penetrating the network, can compromise and leak confidential data (e.g., financial data, medical records, etc.). Federated learning mitigates this concern by focusing on a decentralized computer architecture, which keeps each client’s data independent of and private to outsiders. Individuals can train their own models and contribute to the global model without revealing the contents of the model.