FATE

FATE is a self-hosted federated learning framework for enterprises and institutions that want to train models together while protecting data privacy. It supports collaboration across participating organizations and uses homomorphic encryption and multi-party computation (MPC) for secure computation.

Its algorithms include logistic regression and tree-based methods, as well as deep learning and transfer learning. That makes it relevant to teams building collaborative machine learning systems across several kinds of modeling tasks. The framework also supports model training and prediction.

Deployment can span multiple nodes, with cluster support intended for scale, reliability and manageability. A single-node deployment is available for simple testing. FATE supports Docker deployment, including Docker Compose for clusters, so teams can run the framework on infrastructure they control.

For developers extending a federated learning system, FATE provides standardized algorithm components and APIs designed to connect different computing engines. Its component design also accommodates different scheduling engines. These are useful distinctions for teams that need to integrate federated training into an existing computing environment.

FATE is written primarily in Python and is open source under the Apache 2.0 license. The project is hosted by the Linux Foundation and has a technical charter that defines contribution and oversight responsibilities.