pfl-research: simulation framework for accelerating research in Private Federated Learning
Abstract
Federated learning (FL) is an emerging machine learning (ML) training paradigm where clients own their data and collaborate to train a global model, without revealing any data to the server and other participants. Researchers commonly perform experiments in a simulation environment to quickly iterate on ideas. However, existing open-source tools do not offer the efficiency required to simulate FL on larger and more realistic FL datasets. We introduce pfl-research, a fast, modular, and easy-to-use Python framework for simulating FL. It supports TensorFlow, PyTorch, and non-neural network models, and is tightly integrated with state-of-the-art privacy algorithms. We study the speed of open-source FL frameworks and show that pfl-research is 7-72 faster than alternative open-source frameworks on common cross-device setups. Such speedup will significantly boost the productivity of the FL research community and enable testing hypotheses on realistic FL datasets that were previously too resource intensive. We release a suite of benchmarks that evaluates an algorithm's overall performance on a diverse set of realistic scenarios. The code is available on GitHub at https://github.com/apple/pfl-research.
Cite
@article{arxiv.2404.06430,
title = {pfl-research: simulation framework for accelerating research in Private Federated Learning},
author = {Filip Granqvist and Congzheng Song and Áine Cahill and Rogier van Dalen and Martin Pelikan and Yi Sheng Chan and Xiaojun Feng and Natarajan Krishnaswami and Vojta Jina and Mona Chitnis},
journal= {arXiv preprint arXiv:2404.06430},
year = {2024}
}