Federated learning is a distributed machine learning approach that enables organizations to collaborate on machine learning projects without sharing sensitive data, such as, patient records, financial data, or classified secrets (Sheller MJ, et al., 2020; Sheller MJ, et al., 2019; Yang Y, et al., 2019; McMahan HB, et al., 2016).
The basic premise behind federated learning is that the model moves to meet the data rather than the data moving to meet the model. Therefore, the minimum data movement needed across the federation is solely the model parameters and their updates.
Open Federated Learning (OpenFL) is a Python 3 project developed by Intel Labs and Intel Internet of Things Group.
- OS: Tested on Ubuntu Linux 16.04 and 18.04.
- Python 3.6+ with a Python virtual environment (e.g. conda)
- TensorFlow 2+ or PyTorch 1.6+ (depending on your training requirements). OpenFL is designed to easily support other frameworks as well.