Towards Pre-Training Data Evaluation for Client Selection in Federated Learning

Federated Learning (FL) [4] is a decentralised approach to machine learning that enables multiple devices, known as clients, to collaboratively train a global model without sharing their local data with a central server. Instead of transferring raw data, clients train their models locally and only share model updates with a central aggregator, which combines these updates to create a global model. This approach ensures data privacy and security, as sensitive data remains on the clients’ devices.