Krum, the Library

A research framework for Byzantine-resilient distributed learning.

Krum provides a modular environment for comparing robust aggregation rules and analyzing results. It implements state-of-the-art Gradient Aggregation Rules (GARs) and Byzantine attack strategies, along with simulation frameworks that reproduce key experiments from the literature.

PyPI Version Python Versions License

Quickstart

Reference

License

Krum is open-sourced under the MIT License. See the License file on GitHub for details. Source code and issue tracker live at https://github.com/calicarpa/krum.

Indices and tables